Advanced search
Publication Cover
South African Journal on Human Rights Volume 17, 2001 - Issue 4
Submit an article Journal homepage
40
Views
1
CrossRef citations to date
0
Altmetric
Original Articles

Sub Judice in South Africa: Time for a Change

Graeme Hill Counsel assisting the Solicitor-General of Australia. I wish to thank Stuart Woolman, Iain Currie and Adrienne Stone (and the anonymous SAJHR referee) for their comments on earlier drafts.
Pages 563-593 | Published online: 02 Feb 2017

  • The Constitution of the Republic of South Africa Act 108 of 1996 (‘the Constitution’). Section 16 states that ‘[e]veryone has the right to freedom of expression, which includes (a) freedom of the press and other media’. This article's contention is not new. For example, Gilbert Marcus argues that ‘[p]rejudice to the right to a fair trial is obviously a particularly strong claim and courts are unlikely to be sympathetic to the demands of free expression in this area. Nevertheless, it is questionable whether the current common law test for contempt would survive constitutional challenge’. G Marcus ‘Freedom of Expression Under the Constitution’ (1994) 10 SAJHR 140, 145.
  • Section 35(3) of the Constitution expressly confers on the accused the right to a fair trial, which has been called a ‘general principle of fundamental importance’. S v Ntuli 1996 (1) SA 1207 (CC) 1208.
  • KW Stuart The Newspaperman's Guide to the Law 2 ed (1977) 79. See also the reference to the ‘due administration of justice’ in S v Harber 1988 (3) SA 396 (A) 418. In S v Mamabolo 2001 (3) SA 409 (CC) 427, Kriegler J referred to ‘the proper administration of justice’ as the public interest justifying the ‘scandalizing the court’ aspect of contempt law. While sub judice is a common law doctrine, the constitutional analysis would seem to be the same if freedom of expression were to be limited by statute (see s 8(1) and (3) of the Constitution).
  • S v Van Niekerk 1972 (3) SA 711 (A) 724. The sub judice doctrine applies to civil proceedings as well as criminal proceedings, although contempt of a civil court is not seen as not quite as serious as contempt of a court trying a criminal case (Stuart, note 3 above 83).
  • Van Niekerk (note 4 above) 724. In Harber, Van Heerden JA stated that South African courts ‘have evolved inter alia [that] criterion’ of tendency (note 3 above 420), which seems to suggest that there might be other tests of tendency.
  • See Harber (note 3 above) 418. The Court did not consider the position of a newspaper proprietor, or of radio or television programs (ibid). See further III(a)(iii) (comparison with UK Contempt of Court Act) and V(c) below.
  • Note 4 above.
  • Apparently on the basis that, for judges to deny creditworthiness to evidence ‘irrespective of its intrinsic merits… would be grossly improper’ (ibid 723). In S v Van Staden 1973 (1) 70 (W), however, a contempt conviction for publishing an opinion piece calling for clemency for people convicted of murder was overturned on appeal. Interestingly, the High Court of Australia has laid down a principle similar to that urged by Professor Van Niekerk, by requiring trial judges, in cases where police evidence of a confession is not reliably corroborated, to warn the jury of the dangers of convicting on the basis of this evidence alone. McKinney v The Queen (1991) 171 CLR 468.
  • Note 3 above 425. Despite its apparent strictness, Harber was in fact more protective of freedom of expression than it could have been, because the Court rejected the argument that liability could be imposed on a newspaper without any finding of fault (as had been the case with defamation), although the Court held that a finding of negligence would ground liability. Secondly, the Court also required that the statements in question ‘improperly’ interfere with the administration of justice before any liability would arise. J Burchell ‘Contempt of Court by the Media: Another Opportunity to Extend Press Freedom is Lost’ (1988) 4 SAJHR 375, 382.
  • See S Woolman ‘Civilisation and its Malcontents: President of the Republic of South Africa v South African Rugby Football Union’ (2000) 16 SAJHR 350, 355 (referring to contempt generally); B Van Niekerk The Cloistered Virtue (1987) 103. See also Mamobolo (note 3 above) para 76 per Sachs J.
  • See note 3 above. See also S v Chinamasa 2001 (1) SACR 278 (referred to with approval in Mamabolo para 24) where the Zimbabwe Supreme Court held that the sub judice doctrine was consistent with the guarantee of freedom of expression contained in s 20 of the Zimbabwe Constitution.
  • See Mamabolo (note 3 above) 702 para 32.
  • Ibid para 17. See s 165(2) and (4) of the Constitution.
  • Ibid para 46.
  • Ibid para 75. Sachs J agreed with the orders of the Court. This difference between the judgments of Kriegler J and Sachs J corresponds to the difference between the minority and majority judgments, respectively, of the Ontario Court of Appeal in R v Kopyto (1987) 47 DLR (4th) 213. See Mamabolo (note 3 above) para 35 per Kriegler J, paras 74 and 75 per Sachs J.
  • See note 3 above and accompanying text (purpose of sub judice doctrine).
  • C Cleaver ‘Ruling Without Reasons: Contempt of Court and the Sub Judice Rule’ (1993) 110 SALJ 530, 531–532.
  • Ibid 532. One difficulty with the sub judice doctrine (and its requirement that proceedings be ‘pending’) is that much prejudicial publicity occurs immediately after the commission of the crime, before any proceedings have been brought. See SJ Krause ‘Punishing the Press: Using Contempt of Court to Secure the Right to a Fair Trial’ (1996) 76 BULR 537, 567.
  • See Abolition of Juries Act 34 of 1969. For a discussion of the problems with the jury system in South Africa that led to its abolishment (mainly to do with race), see E Kahn ‘Restore the Jury? Or “Reform? Reform? Aren't Things Bad Enough Already?”’ (1992) 109 SALJ 87, 96–103. Craig Cleaver observes that Van Niekerk does not explain adequately why a doctrine developed primarily to shield juries from irrelevant information should continue to apply in South Africa once juries had been abolished (Cleaver (note 17 above) 532–33).
  • In criminal cases, it seems more likely that the prosecution would attempt to introduce ‘factual’ prejudicial material in court, even if ultimately ruled inadmissible (say, prior convictions, or incriminating physical evidence), than to introduce prejudicial material of a more ‘emotional’ nature (say, a claim that the defendant is a suspect in a hit-and-run killing of a child involving the same motor vehicle used in the robbery). For an explanation of the factual/emotional publicity distinction and the effect of each on conviction rates see CA Studebaker & SD Penrod ‘Pretrial Publicity: The Media, the Law, and Common Sense’ (1997) 3 Psych Pub Pol and L 428, 437, 447–48.
  • Cleaver (note 17 above) 533–34.
  • Note 11 above, 298.
  • Neil Vidmar distinguishes between four different types of prejudice: (i) interest prejudice (having a stake in the outcome), (ii) specific prejudice (having attitudes or beliefs about this particular case), (iii) generic prejudice (having general attitudes or beliefs that would be transferred to this case), and (iv) conformity prejudice (perceiving an overwhelming community interest in a particular outcome). N Vidmar ‘The Canadian Criminal Jury: Searching for a Middle Ground’ (1999) 62 L & Contemp Prob 141, 155–156. Judges may conceivably be better at recognising (and hence overcoming) specific prejudice and conformity prejudice (at least in the case of non-elected judges), but it is less clear whether judges would be any better at identifying generic prejudice.
  • Section 145(2) of the Criminal Procedure Act 51 of 1977 provides that a judge ‘may summon not more than two assessors to assist him at the trial’ in cases when a not guilty plea is entered. This measure was introduced into South Africa in 1917 (see Kahn (note 19 above) 94). The role of assessors is discussed in D van Zyl Smit & N Isakow ‘Assessors and Criminal Justice’ (1985) 1 SAJHR 218.
  • Cleaver (note 17 above) 534, Van Zyl Smit & Isakow (note 24 above) 222.
  • Section 145(1)(b) of the Criminal Procedure Act states only that an assessor is someone who, in the opinion of the presiding judge, ‘has experience in the administration of justice or skill in any matter which may be considered at the trial’.
  • Cleaver (note 17 above) 534; Van Niekerk (note 4 above) 723. Note, however, that s 145(4)(b) of the Criminal Procedure Act allows a judge to exclude the assessor(s) from any consideration whether evidence of any confession or admission by the accused is admissible. Moreover, there is some authority that an assessor should withdraw from hearing a case if she receives information detrimental to the accused which has not been proved in evidence. See R v Matsego 1956 (3) SA 411 (A); R v H 1955 (2) SA 288 (T).
  • Craig Cleaver argues that this need for protection justifies the sub judice rule (Cleaver (note 17 above) 534). He dismisses an argument that sub judice might be required to ensure that improper communications do not deter persons with good claims from commencing action, or induce them to reach a compromise, on the basis that this would be an undue interference with the autonomy of the litigants (ibid 535–36). By contrast, English courts have been prepared to give some weight to this consideration. See, for exmple, Attorney-General v Times Newspapers Ltd [1974] AC 273 (HL) and Attorney-General v Hislop [1991] 1 QB 514 (CA).
  • MR Stabile ‘Free Press-Fair Trial: Can They Be Reconciled in a Highly Publicized Criminal Case?’ (1990) 79 Geo LJ 337, 337.
  • Cleaver (note 17 above) 535, 537.
  • As stated in Brennan J's concurring judgment in Richmond Newspapers Inc v Virginia 448 US 555, 596–597 (1980). Similar arguments arise in relation to the televising of trials: some argue that televising proceedings will make it more likely that witnesses will lie, some argue that it will make it more likely that witnesses will tell the truth. See, for example, GC Harris ‘The Communitarian Function of the Criminal Jury Trial and the Rights of the Accused’ (1995) 74 Neb LR 804, 835–39.
  • See SM Matheson Jr ‘The Prosecutor, the Press and Free Speech’ (1990) 58 Fordham LR 865, 883.
  • In Australia, see the comments of Hayne JA (as he then was) about the ‘incompatibility’ principle set out in Kable v Director of Public Prosecutions (1996) 189 CLR 51 which, broadly, requires judges to assess the affect of legislation on ‘public confidence’ in the judiciary. Hayne JA stated: ‘Is the perception that reasonable members of the public may have of it relevant? If so, what kind of perception is relevant?’ R v Moffatt [1998] 2 VR 229, 251.
  • Matheson (note 32 above) 883, Burchell (note 9 above) 385. Public comment about the judicial process will in many situations actually promote the fairness of a trial (see note 154 below). Craig Cleaver argues that public confidence is irrelevant to the sub judice doctrine, because the doctrine is concerned only with the effect on the impartiality of the proceedings, not the public's perception of that impartiality (Cleaver (note 17 above) 538). On this view, the appropriate means for preserving public confidence in the judiciary is the ‘scandalising the court’ aspect of contempt law. Alternatively, some have argued that any damage to the court's reputation should be dealt with by the law of defamation, rather than by the law of contempt (Burchell (note 9 above) 382–85). As noted in note 150 below, the common law of defamation in South Africa now places more weight on the importance of freedom of expression than formerly, at least where public figures are involved.
  • Note 3 above, para 30 per Kriegler J. See also para 76 per Sachs J, who stated that ‘the more critics were suppressed [under the former application of the “scandalising the court” aspect of contempt], the greater the loss of prestige of the judiciary’.
  • In this sense, there is at least a rational (if not a proportionate) relationship between the purposes of the sub judice doctrine (promoting the due administration of justice) and the restrictions it places on freedom of expression. See s 36(1)(d) of the Constitution and note 123 below.
  • Section 39(1)(c) of the Constitution provides that courts ‘may consider foreign law’ in interpreting the Bill of Rights.
  • Cleaver (note 17 above) 535, AE van Blerk Judge and Be Judged (1988) 5.
  • The UK Contempt of Court Act is not, however, an exhaustive statement of the law restricting publication (which has been referred to as ‘a bewildering hotchpotch of obscure common law, statutory powers and administrative advice’). C Walker, I Cram & D Brogarth ‘The Reporting of Crown Court Proceedings and the Contempt of Court Act 1981’ (1992) 55 Modern LR 647, 666. One of the major reasons for the reforms contained in the UK Contempt of Court Act was the decision of the European Court of Human Rights in The Sunday Times v United Kingdom (1979) 2 EHRR 245, which held that the English common law of contempt was contrary to the guarantee of freedom of expression contained in art 10 of the European Convention on Human Rights. SH Bailey ‘The Contempt of Court Act’ (1982) 45 Modern LR 301, 301.
  • See ss 1 and 2 of the UK Contempt of Court Act. It is a strict liability rule because liability does not depend upon establishing an intention to interfere with the proceedings (si).
  • Section 5 of the UK Contempt of Court Act. This is an exclusion from the strict liability rule, ather than a defence, so it is necessary for the prosecution to prove either that the publication did not contain a good faith discussion of public affairs, or that the risk of impediment or prejudice was not merely incidental. Attorney-General v English [1983] 1 AC 116 (HL).
  • See s 4(1) of the UK Contempt of Court Act, which reflects the common law position (R v Evening News; ex parte Hobbs [1925] 2 KB 158). A court may order that publication of particular proceedings be ‘postponed’ if it appears ‘necessary for avoiding a substantial risk of prejudice to the administration of justice’ (s 4(2)).
  • Section 3(1) of the UK Contempt of Court Act. The requirement that proceedings be ‘active’ is defined in s 2(3) and Schedule 1 to the Act. In general terms, proceedings are ‘active’ once ‘initial steps’ are taken in the prosecution (such as the issue of a warrant for arrest: see c1 4(b)), and the ‘active’ period includes appeals (see ell 15 and 16). There is also a defence for the distributor of a publication that the distributor did not know (having taken reasonable care) or reasonably suspect that the publication contained material in contempt of court (s 3(2)). As s 3(1) and (2) are defences, the onus rests on the publisher or distributor to establish any relevant facts (s 3(3)).
  • As noted above, the South African ‘tendency’ test asks whether, if the facts as set forth in the document were accepted by the tribunal, this would influence the outcome (see note 5 above).
  • No such exception exists in South African common law, unless perhaps it comes within the requirement that the publication is an ‘improper’ interference with the administration of justice (see note 9 above). The English statutory exception is based on the Australian common law decision of Re Truth and Sportsman Ltd; ex parte Bread Manufacturers Ltd (1937) 37 SR (NSW) 242, cited with approval by Lord Reid in Attorney-General v Times Newspapers Ltd [1974] AC 273, 296–297. See also 319, 321. Recognising this exception may well have changed the outcome of Van Niekerk, because arguably any prejudice to the Hassim proceedings was incidental to Professor Van Niekerk's generally phrased comments about the operation of the Terrorism Act. Cf Van Niekerk (note 4 above) 722–23.
  • The Court of Appeal has stated that s 4(2) should be read narrowly (R v Horsham JJ; Ex parte Farquharson [1982] QB 762, 793) and that trial judges should consider whether there are any alternatives before making orders under this provision (Ex parte Central Television Plc [1991] 1 WLR 1, 8–9). See also notes 69 and 91 below (Canadian and American requirement to consider alternative measures to restricting speech).
  • Walker, Cram & Brogarth (note 39 above) 648. Section 4(2) also applies to ‘pending or imminent’ proceedings, which is a broader notion than the ‘active’ proceedings referred to in the strict liability rule (ibid, 649).
  • Section 11 codifies a common law power which was used to protect the names of victims in blackmail cases, in cases of national security, and in other situations where witnesses might be discouraged by a fear of victimisation or lurid publicity. Walker, Cram & Brogarth (note 39 above) 651.
  • For example, some argue that publication of evidence adduced in a voir dire (ie when the jury was excluded) would not be in ‘good faith’, on the basis that ‘reporters should be aware that the information was not intended for the ears of the jurors’. DP Carney ‘The Accused, the Jury and the Media’ (1995) 145 New LJ 12, 25. The exact scope of s 11 (discussed in note 48 above) is also unclear. Walker, Cram & Brogarth (note 39 above) 651–52.
  • See, for example, JA Brandwood ‘You Say “Fair Trial” and I Say” Free Press“: British and American Approaches to Protecting Defendants' Rights in High Profile Trials’ (2000) 75 NYULR 1412, 1435.
  • Alternatively, uncertainty in the law may cause publishers to disregard the law entirely, which may result in trials being prejudiced. S Walker ‘Freedom of Speech and Contempt of Court: The English and Australian Approaches Compared’ (1991) 40 ICLQ 583.
  • Problems of over-inclusion can occur even with what would seem to be clearly prejudicial material, such as the criminal record of a person who the newspaper alleges is the perpetrator, or a report that a person has made a confession to police. Publication of this material may in fact never have any prejudicial effect, because the person may never be charged, or that material may in due course be adduced into evidence. Matheson (note 32 above) 903–04.
  • The UK Contempt of Court Act (like the sub judice doctrine) is concerned with the due administration of justice, rather than the accused's rights as such. Nonetheless, it can have the effect of preserving the accused's right to a fair trial by suppressing the publication of information that would prejudice the trial.
  • See note 43 above.
  • Carney (note 49 above) 12.
  • Brandwood (note 50 above) 1440–1441. Michael Chesterman argues, however, that, in relation to information received via the internet, research is needed on the nature of information appearing there, and the extent to which this information is remembered, believed and further disseminated. M Chesterman ‘OJ and the Dingo: How Media Publicity Relating to Criminal Cases Tried by Jury is Dealt With in Australia and America’ (1997) 45 Am J Comp L 109, 143.
  • In countries with a jury system, the alternative measures are designed mainly to preserve the impartiality of juries (see Brandwood (note 50 above) 1444), a consideration that is of course not relevant to South Africa (see note 19 above).
  • Note 3 above 422.
  • See note 6 above.
  • Cleaver (note 17 above) 540 n48. The issue of mens rea is discussed further in V(c) below.
  • More typically, however, the publisher will have some information and will attempt to draw inferences on whether the proceedings are ‘active’. In this situation, the publisher will have ‘reason to suspect’ that the proceedings are active and the s 3(1) defence will be unavailable. CJ Miller Contempt of Court 2 ed (1989) 289. The defence of being unaware that material was in contempt, having taken reasonable care, is only available to a distributor, not a publisher (s 3(2)).
  • See note 19 above. It should be noted, however, that the definition of ‘active’ proceedings in the UK Contempt of Court Act means that the strict liability rule applies to appellate proceedings (which of course do not involve juries). Cf Bailey (note 39 above) 308.
  • See s 36 of the Constitution, discussed in IV below.
  • See Canadian Charter of Rights and Freedoms, ss 2(b) (‘freedom of… expression, including freedom of the press and other media of communication’) and 11(d) (‘Any person charged with an offence has the right… to be presumed innocent until proven guilty according to law in a fair and public hearing by an independent and impartial tribunal’).
  • Section 1 of the Charter provides that the rights and freedoms set out are subject ‘only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society’.
  • See Stuart (note 3 above) 91–93. This is also the approach taken in Australia. See Chesterman (note 56 above) 116.
  • R Nunez ‘Calibrating the Scales of Justice. Balancing Fundamental Freedoms in the United States and Canada’ (1997) 14 Ariz J Inl & Comp Law 551, 560. Various Canadian statutory provisions that limit freedom of expression are set out in A Riehle ‘Canada's “Barbie and Ken” Murder Case: The Death Knell of Publication Bans?’ (1996) 7 Ind Int & Comp LR 193, 198.
  • [1994] 3 SCR 835, 878.
  • Ibid 881. These alternative measures are also used in the United States (see note 91 below).
  • Ibid 890–891. Lamer CJ also mentioned other ‘general guidelines’: the media should be given standing to argue why a publication ban should not be made, and that the party seeking the ban bears the burden of justifying it (ibid).
  • Ibid 875. The South African tendency test is set out in the text accompanying note 5 above.
  • As noted by Gonthier J in his dissenting judgment (ibid 927–928). The screening of jurors in the 1995 murder trial of Paul Bernado took five days, which was ‘extraordinarily lengthy’ for a Canadian trial. Vidmar (note 23 above) 145.
  • Vidmar (note 23 above) 157–165. These problems are discussed more fully at notes 95–100 below. Note that, in Canada, postponing the trial could potentially conflict with s 11 (b) of the Charter, which requires a trial to be ‘within a reasonable time’.
  • R Martin ‘Judicial Proceedings: Media Bans: CBC v Dagenais’ (1995) 74 Canadian Bar R 500, 509.
  • The second part of the Dagenais test applies the general test for interpreting s 1 of the Charter laid down in R v Oakes [1986] 1 SCR 103 (Dagenais, note 68 above, 887). For an application of the Oakes test in the related context of press access to court proceedings, see Canadian Broadcasting Corp v New Brunswick (Attorney-General) [1996] 3 SCR 480, paras 36–66) (deciding that statutory limitations were reasonable).
  • Jeremy Waldron argues that ‘when people talk about weighing or balancing one value, principle, or consideration against another, what they mean is not necessarily Benthamite quantification but any form of reasoning or argumentation about the values in question’. J Waldron ‘Fake Incommensurability: A Response to Professor Schauer’ (1994) 45 Hastings LJ 813, 819. See also S Woolman Out of Order? ‘Out of Balance? The Limitation Clause of the Final Constitution’ (1997) 13 SAJHR 102, 103 (‘One thing we cannot do easily is balance the incommensurable goods that constitutional cases often require us to compare’).
  • Martin (note 74 above) 508.
  • Cf the limitations of the UK Contempt of Court Act (see notes 52–53 above).
  • Of course, at some point the two rights will conflict, and it will be necessary for the trial judge to decide whether to qualify freedom of expression in order to prevent an unacceptable risk to the fair trial rights of an accused, or whether the potential risk of prejudice to the accused is so small as not to justify limiting freedom of expression.
  • See, for example, R v Banville (1983) 3 CCC (3d) 312, 316 (‘freedom of the press must, in my view, give way to the overriding obligation to ensure that an individual have a fair trial’).
  • In Dagenais, the Supreme Court of Canada altered the common law rules relating to publication bans because, under the Charter, freedom of expression was to be given equal importance as the right to a fair trial. It was therefore necessary to consider under s 1 of the Charter (the equivalent of s 36 of the Constitution) whether a fair trial could be achieved in ways that did not limit freedom of expression (note 68 above, 875–877).
  • See United States Constitution, First Amendment (‘Congress shall make no law… abridging the freedom of speech, or of the press’) and the Sixth Amendment (‘In all criminal prosecutions, the accused shall enjoy the right to a speedy and public trial, by an impartial jury'; see also the due process guaranteed by the Fifth and Fourteenth Amendments).
  • 314 US 252, 261–63 (1941).
  • The Supreme Court has held that trial judges are under a positive obligation to ensure that an accused receives a fair trial. Sheppard v Maxwell 384 US 333, 363 (1966): ‘The courts must take such steps… that will protect their processes from prejudicial outside influences’. Some argue that this statement overlooks the requirement of ‘State action’, and therefore that the accused may only prevent the state, and not the press, from acting unconstitutionally. HA Linde ‘Fair Trials and Press Freedom — Two Rights Against the State’ (1977) 13 Williamette LJ 211. Cf the discussion of state action and inaction in DeShaney v Winnebago County Department of Social Services 489 US 189 (1989).
  • Rodney Smolla argues that the United States ‘stands out in the world as one of the great free speech “laboratories of experiment”’ (RA Smolla Free Speech in an Open Society (1993) 351). For example, courts in many countries (including South Africa) have considered the rule in New York Times v Sullivan 376 US 254 (1964), even if none have accepted that rule without modification. See, for example, A Stone & G Williams ‘Freedom of Speech and Defamation: Developments in the Common Law World’ (2000) 26 Monash LR 362.
  • 427 US 539, 562 (1976).
  • Chief Justice Burger stated that ‘pre-trial publicity — even pervasive, adverse publicity — does not inevitably lead to an unfair trial’ (ibid 554), and held that ‘it is not clear that further publicity, unchecked, would so distort the views of potential jurors that 12 could not be found who would, under proper instructions,… render a just verdict exclusively on the evidence presented in open court’ (ibid 569).
  • Ibid 569, see also 565. This onerous requirement is a product of American suspicion of prior restraints on speech, exemplified in cases like Near v Minnesola 283 US 697 (1931). See Nebraska Press (note 86 above) 559. Critics argue that the failure to make the ban on restraining orders absolute (which Burger CJ refused to do: ibid 569–570) has simply shifted the argument to whether the trial judge should restrain speech while she determines whether the publication in question is sufficiently exceptional to justify issuing a restraining order. A Bernabe-Riefkohl ‘Prior Restraints on the Media and the Right to a Fair Trial: A Proposal for a New Standard’ (1996) 84 Ky LJ 259, 284.
  • See, for example, Bernabe-Riefkohl (note 88 above) 290–293. Indeed, Nebraska Press seems to impose a Catch-22 situation on speech limitations: as ‘it would be speculative to conclude that any [alternative] measure would have failed, the Court [in Nebraska Press] must have meant that the alternatives should be tried before any restraint is imposed’. Matheson (note 32 above) 927. Yet if these measures have proved to be ineffective, freedom of speech cannot be limited then either, because the limitation would be futile. See the third prong of the Nebraska Press test set out above.
  • See note 69 above.
  • Nebraska Press (note 86 above) 563–564.
  • In Gentile v State Bar of Nevada 501 US 1030 (1991) the Supreme Court (by a 5:4 majority) upheld a State bar rule that prohibited a lawyer from making statements that the lawyer ‘knows or reasonably should know will have a substantial likelihood of materially prejudicing an adjudicative proceeding’. The majority considered that the speech of lawyers could be regulated under a less demanding standard than that applicable to regulation of the press, because of a lawyer's duties to the court (1074). However, the fact that Gentile concerned subsequent punishment of speech, rather than a prior restraint, has meant that some commentators have argued that ‘gag orders’ imposed on lawyers are unconstitutional. See, for example, E Chemerinsky “Lawyers Have Free Speech Rights Too: Why Gag Orders on Trial Participants Are Almost Always Unconstitutional’ (1997) 17 Loy LA Ent LJ 311.
  • See the discussion of Californian legislation prohibiting potential witnesses from selling their stories to the media, and New York legislation prohibiting criminals from profiting from a crime before the victims are compensated, in MJ Conboy & AR Scott ‘Tipping the Scales of Justice: An Attempt to Balance the Right to a Fair Trial with the Right to Free Speech’ (1996) 11 Si Johns JL Comm 775, 796–98.
  • In Nebraska Press, Burger CJ stated ‘we cannot ignore the reality of the problems of managing and enforcing pre-trial restraining orders' (note 86 above 565).
  • Studebaker & Penrod (note 20 above) 442–43, Krause (note 18 above) 565. Studebaker and Penrod suggest, however, that judicial instructions may be somewhat effective in getting jurors to disregard publicity that occurs after the trial has begun, provided this warning is given before the trial starts (note 20 above 444).
  • Studies suggest that potential jurors are unlikely to identify prejudice in themselves or likely to over-estimate their ability to decide only on the basis of adduced evidence, and further that lawyers and judges are ineffective in identifying prejudice in potential jurors. Studebaker & Penrod (note 20 above) 440–43. Moreover, as a practical matter, it is difficult to ask potential jurors about prejudicial material without revealing that material. Ibid 443. See also Brandwood (note 50 above) 1444.
  • For example, critics of ‘Mu'Min v Virginia 500 US 415 (1991) argue that it ‘requires [the accused] to demonstrate that his jury was biased, but fails to guarantee him the procedural device necessary to meet that burden’. BP Coffey ‘Mu'min v Virginia: Reexamining the Need for Content Questioning During Voir Dire in High Profile Criminal Cases’ (1993) 13 Pace LR 605, 638.
  • See, for example, Krause (note 18 above) 565, Stabile (note 29 above) 343–44.
  • Krause (note 18 above) 566. Stephen Krause notes further that a change of venue may be ‘dramatically less effective, if not completely useless’ in cases that receive national publicity (ibid). Even in cases of national publicity, however, a change of venue may reduce jury prejudice, without eliminating it. See Studebaker & Penrod (note 20 above) 450–53 discussing the change of venue in the Timothy McVeigh trial.
  • See, for example. Stabile (note 29 above) 345. Studebaker and Penrod suggest that postponement may be more effective in dispelling prejudice caused by ‘factual’ publicity than prejudice caused by ‘emotional’ publicity (note 20 above 440).
  • See, for example, Brandwood (note 50 above) (recommending limitations on lawyer speech); Krause (note 18 above) 566 (recommending subsequent punishment of the media using the contempt power).
  • Michael Chesterman argues that consideration could be given to ‘mixed’ juries, ie comprising both lay and judicial members. Chesterman (note 56 above) 147. Historically, of course, jurors were chosen precisely because of their knowledge of the case. NN Minow & FH Cate ‘Who is an Impartial Juror in an Age of Mass Media?’ (1991) 40 Am Univ LR 631, 637–38.
  • Some discount the empirical evidence of jury bias entirely. See, for example, RM Jones ‘The Latest Empirical Studies on Pretrial Publicity, Jury Bias, and Judicial Remedies — Not Enough to Overcome the First Amendment Right of Access to Pretrial Hearings' (1991) 40 Am Univ LR 841. For a discussion of the gaps in the existing empirical evidence, see Studebaker & Penrod (note 20 above) 446–47.
  • See, for example, RD Sack ‘Principle and Nebraska Press Association v Stuart’ (1977) 29 Stan LR 411, 416, who states that ‘protection for the press must be strict and unyielding’. See also Bernabe-Riefkohl (note 88 above) 316.
  • Some commentators dismiss the distinction between orders restraining the ability of the media to disseminate information, and orders limiting the ability of the media to gather information, as ‘formalistic’. See, for example, SA Bjork ‘Indirect Gag Orders and the Doctrine of Prior Restraint’ (1989) 44 Univ Miami LR 165, 181–82. Other commentators, however, accept the validity of this distinction. See Sack (note 104 above) 427.
  • Mamabolo (note 3 above) para 41 per Kriegler J. The United States Supreme Court views freedom of expression as ‘the matrix, the indispensable condition of nearly every other form of freedom’. Palko v Connecticut 302 US 319, 327 (1937).
  • Cf Mamabolo (note 3 above) para 41 per Kriegler J.
  • Ibid para 40 per Kriegler J.
  • In Bridges v California, Black J stated that ‘one of the objects of the Revolution was to get rid of the English common law on liberty of speech and of the press' (note 83 above 264).
  • See Mamabolo (note 3 above) para 37 per Kriegler J. Mahomed DP has stated that the new constitutional order ‘constitutes a decisive break from a culture of apartheid and racism to a culture of openness and democracy and universal human rights' (Shabalala? Attorney- General of Transvaal 1996 (1) SA 725 (CC) para 26). This is not to deny that there are significant historical differences between the situation in the United States in 1789 and in South Africa in 1996 (or 1993 for that matter), so that structures and doctrines that are appropriate for the United States are not necessarily appropriate for South Africa. See, for example, J Zasloff ‘The Tyranny of Madison’ (1997) 44 UCLA LR 795 (arguing that a federal structure is inappropriate for South Africa).
  • See s 16(2)(c) of the Constitution. Cf RAV v City of St Paul 505 US 377 (1992) (attempt to penalise cross-burning was unconstitutional viewpoint discrimination).
  • Michael Chesterman argues that this unwillingness underlies Nebraska Press. See Chester-man (note 56 above) 139.
  • See II(b)(i) above.
  • See note 10 above (sub judice doctrine used to suppress speech critical of government). In fact, the Constitutional Court's approach in Mamabolo (note 3 above) suggests that it is unlikely that South African courts will be as strongly protective of freedom of expression as American courts, even in this context (see V(a) below).
  • The common law has long recognised the importance of freedom of expression. See, for example, Du Plessis v De Klerk 1996 (3) SA 850 (CC) para 58. In the case of sub judice, however, this recognition did not translate into effective protection of the freedom of expression. The Bill of Rights applies expressly to common law rules (such as sub judice), and is not limited to legislative and executive action (see s 8(1) and (3) of the Constitution).
  • The sub judice doctrine is clearly a law of general application.
  • Section 16(2) states ‘[t]he right in subsection (1) does not extend to — (a) propaganda for war; (b) incitement of imminent violence; or (c) advocacy of hatred that is based on race, ethnicity, gender or religion, and that constitutes incitement to cause harm’.
  • Clearly, the values to be served by a right are crucial in determining whether any limits on this right are ‘reasonable and justifiable’ for the purposes of s 36. In some situations (eg determining whether something is ‘expression’), the values to be served by a right will also be relevant in determining whether there has even been a limit on this right to begin with. See S Woolman ‘The Right Consistency: Beinash v Ernst & Young’ (1999) 15 SAJHR 166, 170 (‘the first step in fundamental rights analysis requires a value-based determination of the ambit of the right itself’).
  • One of the various reasons given for providing constitutional protection to freedom of expression is because it promotes a democratic system of government, by giving citizens access to information and to different viewpoints. See, for example, J de Waal, I Currie & G Erasmus The Bill of Rights Handbook 2 ed (1999) 298. See also Chinamasa (note 11 above) 290.
  • In Chinamasa, the Zimbabwe Supreme Court concluded that ‘statements made on a matter of public interest [in that case, criticism levelled at the courts] even if intemperately or offensively worded, or in fact false, so long as they are not obscene or criminally defamatory, come within the protection of s 20(1) of the [Zimbabwe] Constitution’ (note 11 above 291).
  • Section 36(1)(a) of the Constitution (‘nature of the right’). The Constitutional Court has stated that ‘[f]reedom of expression lies at the heart of a democracy’. South African National Defence Union v Minister of Defence 1997 (4) SA 469 (CC) para 7.
  • Section 36(1)(b) (‘importance of the purpose of the limitation’). In the case of criminal proceedings, the Constitutional Court has referred to the ‘great public interest in ensuring that justice is manifestly done to all’. Key v Attorney-General, Cape Provincial Division 1996 (4) SA 187 (CC) para 13.
  • Section 36(1)(d) (‘the relation between the limitation and its purpose’), which requires only a rational relationship between means and end. See, for example, Woolman (note 76 above) 110. In Chinamasa, Gubbay CJ stated that the restrictions on freedom of expression required by the ‘scandalising the court’ aspect of contempt were ‘unquestionably’ for the purpose of protecting the administration of justice (note 11 above 292).
  • Section 36(1)(e) (‘less restrictive means to achieve the purpose’). The inquiry here does not seem to be whether there are any alternative measures, no matter how far-fetched. Accordingly, the narrow tailoring enquiry under s 36(1)(e) is merely a specific aspect of the larger question imposed by s 36(1)(c), namely, whether the purpose is worth the restrictions imposed. See, in the Australian context, A Stone ‘The Limits of Constitutional Text and Structure: Standards of Review and the Freedom of Political Communication’ (1999) 23 Melb Univ LR 668, 682.
  • Section 36(1)(c) (‘the nature and extent of the limitation’). This factor invites a cost-benefit analysis, and requires the courts to ‘revisit the compromise of social interests struck by the co-ordinate branches of government’. Woolman (note 76 above) 110, 111. There is less of a ‘counter-majoritarian difficulty’ where (as here) judges are reviewing the balance struck at common law between two rights (as opposed to judges reviewing the balance struck by legislation).
  • Mamabolo (note 3 above) para 49.
  • These jury-specific measures include sequestration, ‘screening’ potential jurors, and providing directions to the jury (see notes 69 and 91 above).
  • Trial postponement is sometimes used in Australia, which (like South Africa) relies mainly on deterrence through penal sanctions rather than preventative measures. Chesterman (note 56 above) 116.
  • For example, changing venue is less likely to be successful with a case of national, as opposed to local, notoriety (see note 99 above).
  • Section 35(3)(d) of the Constitution; cf notes 73 and 100 above (comparable argument in Canada and the United States).
  • See notes 92–93 and 105 above (discussion of United States position).
  • Of course, regulating speech on the basis of the identity of the speaker is often highly problematic, and can amount to what the Americans term ‘view-point’ discrimination. See K Greenawalt ‘Viewpoints from Olympus’ (1996) 96 Colum LR 697, 700. It is less problematic here when the persons subject to greater restrictions represent the state (as opposed to persons critical of the state), and when the reason for imposing the restrictions is to uphold an important competing private interest (the fair trial rights of the accused).
  • Section 165(3) also provides that ‘[n]o person or organ of the state may interfere with the functioning of the courts’. In Mamabolo, Kriegler J relied on the reference in s 165 to the ‘dignity’ of the courts as one reason to maintain the offence of ‘scandalising the court’ (see note 13 above and accompanying text).
  • Matheson (note 32 above) 890. Accordingly, imposing restrictions on the speech of prosecutors can be an effective way of preventing prejudicial information from being published. To be effective, however, the rules would need to impose an obligation on the prosecutor to ensure that other government officials (such as police officers) do not disclose prejudicial information either (891–892). It may also be necessary to guard against prejudicial material being included in documents filed in court (which would ordinarily be available to the public) (891). See also Brandwood (note 50 above) 1450.
  • In Gentile, Kennedy J considered that the speech in issue was ‘classic political speech’, because it involved speech by defence counsel ‘critical of the government and its officials’ (note 92 above 1034). Some commentators argue, however, that criminal defendants, as a class, will be better off if defence counsel are not expected to vouch for their client outside the courtroom, because many outstanding lawyers may be unwilling to appear to lend their personal reputation and credibility to the client's cause. DA Strauss ‘Why It's Not Free Speech Versus Fair Trial’ 1998 U Chi Legal F 109, 110 (1998).
  • One gap in the sub judice doctrine is that it does not catch conduct that occurs before proceedings are ‘pending’, even though this conduct can be prejudicial to the accused (see note 18). See also note 53 above (discussing the position under the UK Contempt of Court Act). Once proceedings have started, however, it may be more appropriate for the trial judge to make specific orders, tailored to the circumstances of that case. Matheson (note 32 above) 895.
  • See note 88 above. See also Republic of South Africa v The Sunday Times Newspaper 1995 (2) BCLR 182 (T).
  • As claimed by Alexander Bickel in The Morality of Consent (1975) 61. Martin Redish, for example, responds that this proposition contains a ‘fundamental fallacy’, because ‘the speech which the prior restraint [validly] keeps from the marketplace of ideas is speech which would not be found constitutionally protected in a subsequent prosecution’. M Redish ‘The Proper Role of the Prior Restraint Doctrine in First Amendment Theory’ (1984) 70 Va LR 53, 59. See also JC Jeffries Jr ‘Rethinking Prior Restraint’ (1983) 92 Yale LJ 409, 429. When prior restraints are imposed by courts, the most significant difference between prior restraints and subsequent punishment is that someone who disobeys the prior restraint cannot challenge the constitutional validity of the restraint in contempt proceedings (the ‘collateral bar’ rule). Rather than being a reason for treating prior restraints differently from subsequent punishment, however, this result may be reason to alter the collateral bar rule. See eg RE Labunski ‘The “Collateral Bar” Rule and the First Amendment: The Constitutionality of Enforcing Unconstitutional Orders’ (1988) 37 Am Univ LR 323.
  • Several American commentators emphasise that the existing ‘totality of the circumstances’ test of whether trials are unfair makes it very difficult for trial judges to know what steps they should take. See, for example, R Hardaway & DB Tumminello ‘Pretrial Publicity in Criminal Cases of National Notoriety: Constructing a Remedy for the Remediless Wrong’ (1996) 46 Am Univ LR 39. Moreover, commentators argue that the fact that, as jurors are not known (and hence cannot be removed from prejudicial publicity) during the investigative stage of a prosecution, this justifies employing a broader standard for permissible limitations on speech than the (narrow) standard that applies once a jury has been empanelled. See, for example, BH Fleming ‘First Amendment Right of Access to Pretrial Proceedings in Criminal Cases' (1983) 32 Emory LJ 619, 670–671. On the other hand, the need for content limitations on speech to be narrowly tailored would ordinarily require case-by-case consideration of whether speech caused prejudice to proceedings (676–677). Moreover, as s 5 of the UK Contempt of Court Act recognises, the commencement of a prosecution cannot totally bar discussion of all matters potentially bearing on future criminal proceedings. One way of resolving the tension between these competing considerations is to prohibit during the early stages of the prosecution the publication of specific categories of information that are likely to be especially prejudicial (even though it may turn out that publishing would not have actually caused prejudice), and not to prohibit other publication of other information during this period (despite the more remote possibility that this more general information could cause prejudice). See Matheson (note 32 above) 893–924.
  • See II(b)(i) above (judge already exposed to inadmissible material).
  • See s 36(1)(e) of the Constitution.
  • See s 36(1)(c) of the Constitution. This is essentially the second question posed by the Dagenais test (see note 68 above).
  • See notes 31 and 34 above.
  • See ss 1(d) and 36(1) of the Constitution, which include accountable democratic government as one of the founding values of the Constitution. See also s 16(1)(b) (freedom of expression includes ‘freedom to receive or impart information or ideas’) and s 32(1) (right of access to information).
  • Holomisa v Argus Newspapers Ltd 1996 (2) SA 588 (W) 609, quoted with approval by Hefer JA in National Media Ltd v Bogoshi 1998 (4) SA 1196 (SCA) 1217. Hefer JA went on to say, however, that he did not think that freedom of expression relating to political activity could be given paramountcy ‘indiscriminately and irrespective of the circumstances of each case’ (1217).
  • V Blasi ‘Learned Hand and the Self-Government Theory of the First Amendment: Masses Publishing Co v Patten’ (1960) 61 Univ Colo LR 1, 13. On one view, certain harms must therefore be tolerated, in order to ensure that freedom of expression can perform this legitimating function. See V Blasi ‘Learned Hand and the Self-Government Theory of the First Amendment: Masses Publishing Co v Patten’ (1990) 61 Univ Colo LR 1, 13.
  • Note 3 above paras 29–31 per Kriegler J.
  • Note 121 above.
  • Ibid para 8 per O'Regan J.
  • Note 145 above 1209–1210 per Hefer JA, stating that ‘it is the right, and indeed a vital function, of the press to make available to the community information and criticism about every aspect of public, political, social and economic activity’, and that the press ‘often becomes the voice of the people — their means to convey their concerns to their fellow citizens, to officialdom and to government’.
  • Stuart Woolman argues that an increased tolerance for criticism also emerges from the Constitutional Court's decision in President of the Republic of South African v South African Rugby Football Union 1999 (4) SA 147 (CC). See Woolman (note 10 above).
  • Note that the speech at issue in both Van Niekerk and Harber was critical of the then government (see notes 8 and 9 above).
  • Section 35(3) of the Constitution. This right ‘embraces a concept of substantive fairness which is not equated with what might have passed muster in our criminal courts before the Constitution came into force’. S v Zuma 1995 (2) 642 (CC) para 16, interpreting s 25(3) of the Constitution of the Republic of South Africa, Act 200 of 1993.
  • Sheppard v Maxwell (note 84 above) 350 per Clark J. See also Stuart (note 3 above) 84.
  • The trial considered in Sheppard v Maxwell (note 84 above) provides just one example. In South Africa, the chances of adverse publicity actually causing prejudice to the accused is much less than in countries that use juries in criminal trials (see II(b)(i) above).
  • See Shabalala (note 110 above) para 35.
  • Vince Blasi has observed that ‘one reason to disfavor laws that apply selectively based on speech content [such as the sub judice doctrine] is that such laws are more likely to have a disparate impact ideological impact than is true for laws that operate irrespective of the content of the speech’. V Blasi ‘Free Speech and the Widening Gyre of Fund-Raising: Why Campaign Spending Limits May Not Violate the First Amendment After All’ (1994) 94 Colum LR 1281, 1292.
  • See note 29 above. Further, English courts at least accept the argument that the sub judice doctrine ensures that potential litigants with good claims are not deterred from bringing legal proceedings (note 28 above).
  • See note 62 above.
  • It seems appropriate for the courts to begin by making modest changes to the common law, rather than attempting straight away to develop a constitutional rule protecting freedom of expression. (Section 173 of the Constitution enables the Constitutional Court to develop the common law.) On arguments in the Australian context why courts should change the common law instead of making constitutional law rules, see A Stone ‘Freedom of Political Communication, the Constitution and the Common Law’ (1998) 26 Fed LR 219. For arguments in favour of judicial minimalism generally see, for example, I Currie ‘Judicious Avoidance’ (1999) 15 SAJHR 138.
  • In its current form, as noted above, the ‘tendency’ test does not require a risk of actual prejudice (see note 5 above).
  • See note 40 above.
  • See note 68 above. The Zimbabwe Supreme Court appeared to accept a similar test in relation to ‘scandalising the court’ contempt in Chinamasa (note 11 above 295).
  • See note 83 above.
  • Mamabolo (note 3 above) para 44 per Kriegler J. In England, the reference in s 2(2) of the UK Contempt of Court Act to ‘substantial’ risk has been interpreted to mean ‘not minimal’ rather than ‘weighty’. Attorney-General v News Group Newspapers Ltd [1987] QB 1, 15. In Canada, however, it is necessary to show that the effect of publicity ‘will be to leave potential jurors so irreparably prejudiced or to so impair the presumption of innocence that a fair trial is impossible’ (Phillips v Nova Scotia (Commission of Inquiry into Westray Mine Tragedy) [1995] 2 SCR 97 para 128).
  • Note 3 above para 43.
  • See note 14 above.
  • Although Sachs J was careful to state that his qualification to the Court's judgment in Mamabolo ‘relates only to this particular class of utterances [statements made outside of court and not relating to ongoing proceedings], and not to the constitutionality of contempt proceedings in general’ (note 3 above para 72).
  • See note 15 above.
  • Cf Mamabolo (note 3 above) para 45 per Kreigler J. However, the requirements of ‘substantial’ risk of ‘serious’ prejudice, if interpreted too strictly, might also lead to the seemingly undesirable result that the courts cannot restrict freedom of expression even in cases where there is a significant (but not substantial) possibility of prejudice to the accused, and alternative measures will not prevent this prejudice.
  • For example, the American and Canadian tests for making orders restraining the media are very similar (see notes 68 and 86 above), but American courts require evidence, not speculation (see note 88 above), and Canadian courts require only ‘satisfactory proof of a link between the publicity and its effects’. Phillips (note 165 above) para 129.
  • Phillips (note 165 above) para 129. Of course, one difference is that Canadian and American courts tend to be assessing the potential prejudice of a publication, whereas South African courts tend to be assessing any prejudice after the fact. However, while it might be thought that the Canadian and American enquiry ‘is necessarily more speculative’ (para 127), some commentators argue that ‘[d]etermining possible prejudicial impact [of publicity] remains just as challenging after the speech has occurred’. Matheson (note 32 above) 896.
  • In MacDonnell v Flahiff, the Quebec Court of Appeal ordered a publication ban even though the applicants did not lead any specific evidence that the publication of the material in question would prejudice their trial ((1998) 17 CR (5th) 94 para 33). The Court was prepared to presume prejudice to the proceedings on the basis that publication of this (admittedly adverse) material would disturb the ‘impartiality and serenity’ required by a fair trial (para 39).
  • In Canadian Broadcasting Corporation v New Brunswick, the Canadian Supreme Court held that ‘[t]here must be a sufficient evidentiary basis from which the trial judge may assess the application and upon which he or she may exercise his or her discretion judicially’ (note 75 above para 72).
  • See note 6 above.
  • See A Arlidge & D Eady The Law of Contempt (1982) para 2–80.
  • A helpful discussion is contained in Miller (note 61 above) Chapter 9 (‘The Sub Judice Rule: Mens Rea and the Scope of Responsibility’).
  • In the United States, it seems that the imposition of strict liability by a State law does not, in itself, contravene the due process clause of the Fourteenth Amendment. See, for example, Bridgeford v U-Haul Company 238 NW 2d 443 at 447 (Supreme Court of Nebraska, 1976)).
  • See note 40 above.
  • See, for example, Arlidge & Eady (note 176 above) 99–100, 165–66. At common law, it was only necessary to show that the accused intended to publish the statements in question. See, for example, Miller (note 61 above) 280.
  • See Re CHBC Television (1999) 23 CR (5th) 135 (British Columbia Court of Appeal) 151–164. In Canada, an offence of ‘absolute liability’ (ie no due diligence defence) is not likely to offend s 7 of the Canadian Charter unless a prison sanction is provided. See Re BC Motor Vehicle Act [1985] 2 SCR 486, as explained by Cory J in R v Pontes [1995] 3 SCR 44 para 26.
  • This and other arguments are set out in Australian Law Reform Commission Report No 35 Contempt (1987) 142–144, para 259.
  • Note 3 above 418.
  • Shabalala (note 110 above) para 26.
  • See notes 34, 146 and 154 above.
  • A judge who takes a strong commitment to having ‘uninhibited, robust and wide-open’ debate on issues of government (New York Times v Sullivan (note 85 above) 270) is likely to take a very different view of the extent of permissible restraints on discussion of pending judicial proceedings than a judge who has more faith in the ability of courts to distinguish between limitations on speech that promote public debate and limitations that do not.

Reprints and Corporate Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

To request a reprint or corporate permissions for this article, please click on the relevant link below:

Order Reprints Request Corporate Permissions

Academic Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

Obtain permissions instantly via Rightslink by clicking on the button below:

Request Academic Permissions

If you are unable to obtain permissions via Rightslink, please complete and submit this Permissions form. For more information, please visit our Permissions help page.

Related research

People also read lists articles that other readers of this article have read.

Recommended articles lists articles that we recommend and is powered by our AI driven recommendation engine.

Cited by lists all citing articles based on Crossref citations.
Articles with the Crossref icon will open in a new tab.