Prospective jurors can safely be told anything about an accused's past. That radical proposition receives some support from the High Court of Australia in today's decision in Dupas v R  HCA 20 (16 June 2010).
The fact that the accused had been previously convicted of two murders that happened in 1997 and 1999 was widely publicised in the Australian media, so that on his present trial for a third murder it was submitted on his behalf that the proceedings should be stayed because he could not get a fair trial.
Untenable as a general proposition though that may be, the question of whether a stay is required must be addressed in the circumstances of each case. When this is looked at from the post-trial appeal perspective, the appellate court can examine the measures that were taken to give effect to the accused's right to a fair hearing.
A general proposition about stays that is usually advanced is that they are only given in extreme cases. The High Court held (18) that the following should be regarded as an authoritative statement of principle:
"[A] permanent stay will only be ordered in an extreme case [citing
Jago v District Court (NSW)  HCA 46; (1989) 168 CLR 23 at 34] and there must be a fundamental defect 'of such a nature that nothing that a trial judge can do in the conduct of the trial can relieve against its unfair consequences' [citing Barton v The Queen  HCA 48; (1980) 147 CLR 75 at 111 per Wilson J;  HCA 48]. And a court of criminal appeal, before it will set aside a conviction on the ground of a miscarriage of justice, requires to be satisfied that there is a serious risk that the pre-trial publicity has deprived the accused of a fair trial. It will determine that question in the light of the evidence as it stands at the time of the trial and in the light of the way in which the trial was conducted, including the steps taken by the trial judge with a view to ensuring a fair trial." R v Glennon (1992) 173 CLR 592;  HCA 16 at pp 605-606 per Mason CJ and Toohey J.
The power to stay proceedings, described as an "inherent power" by Lord Blackburn in Metropolitan Bank Ltd v Pooley (1885) 10 App Cas 210 at 220-221, has what is broadly called a constitutional dimension, which makes the courts reluctant to interfere with decisions whether to prosecute or to continue with prosecutions: Moevao v Department of Labour  1 NZLR 464 at 481, and Fox v Attorney-General  3 NZLR 62 at para 28.
Thus the obvious common-sense objections to allowing stays to recidivists whose records are publicised are supported by these strains of legal principle.
The law must assume that juries will obey instructions given to them by trial judges. The High Court noted this important point as a counter to the denial of the "social imperative" that requires that an accused be brought to trial, that would occur if the media were able to render an accused unable to be tried. The Court referred (26) to the remarks of Hughes J, endorsed by the Court of Appeal, in R v Abu Hamza  QB 659 at 685-686:
"Extensive publicity and campaigns against potential defendants are by no means unknown in cases of notoriety. Whilst the law of contempt operates to minimise it, it is not always avoidable, especially where intense public concern arises about a particular crime and a particular defendant before any charge is brought. Jurors are in such cases capable of understanding that comment in the media might or might not be justified and that it is to find out whether it is that is one of their tasks. They are capable of understanding that allegations which have been made may be true or may not be and that they, the jury, are to have the opportunity and responsibility of hearing all the evidence which commentators in the media have not and of deciding whether in fact the allegations are true or not. They are not surprised to be warned not to take at face value what appears in the media, nor are they these days so deferential to politicians as to be incapable of understanding that they should make no assumptions about whether any statements made by such people are justified or not. They are also capable of understanding and habitually apply the direction that they are given about the standard of proof."
True, there may sometimes be perverse verdicts favouring the accused (27):
" ... in Gammage v The Queen [ HCA 68; (1969) 122 CLR 444 at 463;  HCA 68] Windeyer J expressed the governing principle in terms which acknowledged that the jury room might not be a place of undeviating intellectual and logical rigour (a point made by Callinan J in Gilbert v The Queen [ HCA 15; (2000) 201 CLR 414 at 440 ;  HCA 15]) by saying:
"A jury in a criminal case may sometimes, from compassion or prejudice or other ulterior motive, fail to perform their sworn duty to determine the case before them according to the evidence. If they do so in favour of the prisoner, and not of the Crown, the law is powerless to correct their dereliction. They must be assumed to have been faithful to their duty. Their verdict must be accepted."
It comes down to this (29):
"What, however, is vital to the criminal justice system is the capacity of jurors, when properly directed by trial judges, to decide cases in accordance with the law, that is, by reference only to admissible evidence led in court and relevant submissions, uninfluenced by extraneous considerations. That capacity is critical to ensuring that criminal proceedings are fair to an accused."
And the point is not so much the extreme or rare nature of the circumstances needed to support a stay, as the assurance of trial fairness (35):
"Characterising a case as extreme or singular is to recognise the rarity of a situation in which the unfair consequences of an apprehended defect in a trial cannot be relieved against by the trial judge during the course of a trial. There is no definitive category of extreme cases in which a permanent stay of criminal proceedings will be ordered. In seeking to apply the relevant principle in Glennon, the question to be asked in any given case is not so much whether the case can be characterised as extreme, or singular, but rather, whether an apprehended defect in a trial is "of such a nature that nothing that a trial judge can do in the conduct of the trial can relieve against its unfair consequences."
There is a risk of misinterpretation of the last sentence in para 37 of the High Court's judgment. The para reads:
"A further consideration is the need to take into account the substantial public interest of the community in having those who are charged with criminal offences brought to trial [footnote: R v Glennon  HCA 16; (1992) 173 CLR 592 at 598 per Mason CJ and Toohey J], the "social imperative" as Nettle JA called it, as a permanent stay is tantamount to a continuing immunity from prosecution [footnote: R v Glennon  HCA 16; (1992) 173 CLR 592 at 599 per Mason CJ and Toohey J]. Because of this public interest, fairness to the accused is not the only consideration bearing on a court's decision as to whether a trial should proceed [footnote: Jago v District Court (NSW)  HCA 46; (1989) 168 CLR 23 at 33 per Mason CJ]."
The dicta on this in Jago precede the common law's recognition of the accused's right to a fair trial as an absolute right which is not to be reduced by being balanced against other rights (see my comments on R(Ullah) v Special Adjudicator 3 September 2004; Grant v R (Jamaica) 20 January 2006; Williams v R (Jamaica) 26 April 2006; Condon v R 24 August 2006; R v Davis 19 June 2008; Al-Khawaja and Tahery v United Kingdom 27 January 2009; R v Grant 18 July 2009). It seems likely that what the High Court means here is that the "social imperative" will be given as full effect as is possible, consistent with the absolute right of the accused to a fair trial.
In the present case there was held to have been no unacceptable risk of trial unfairness and the appeal was dismissed. This case did not require examination of the full extent of the inherent power to stay proceedings.
This case illustrates that jury trials can be sufficiently robust to function correctly in the context of vigorously exercised freedom of speech.