Editorial: Jailhouse secrets need to be tested

Defence lawyers routinely warn high-profile clients not to talk about their case to other prison inmates. Yet somehow cell-mates seem, just as routinely, to keep coming forward to testify for the prosecution. The trials of the pair accused of Peter Plumley-Walker's murder, and of David Tamihere, both involved jailhouse witnesses. So did the Scott Watson double-murder trial.

Two inmates who spent time with Watson on remand became important witnesses for the Crown when they alleged he had confessed to the Sounds murders. Now one of those men, Witness A, says the police pressured him into giving false evidence. Never has the unreliability of such witnesses been so graphically illustrated.

The judge in the Watson trial went quite some way to warn jurors on that very point. In his summing-up, Justice Heron asked the jury to consider whether it was likely that Watson would have confided in Witness A when he had not confided in his girlfriend. Of Witness B, he noted that the man had a purpose of his own to serve - his charges had been reduced, among other things - and that jurors should be cautious about accepting the testimony.

There are other, more fundamental, objections to the use of such witnesses. Crown lawyers, along with the police, may say they are uncomfortable with this type of testimony. But it seems only human nature for the Crown to use such witnesses when its chances of a successful prosection are borderline.

Judges undoubtedly also feel uncomfortable about such testimony, but in most cases they seem content to allow the evidence of jailhouse witnesses to go before the jury. It is not, they might contend, their job to determine truthfulness; that is the jury's core function. Only in extreme cases do judges seem willing to conclude that a witness' credibility is so tainted and compromised that it would be unfair to the accused to allow the testimony.

Witness A's dramatic disclosure suggests that that relatively passive course may no longer suffice. No one can know how much weight the Watson trial jury placed on the evidence of Witnesses A and B. Yet of the almost 500 witnesses, only they presented graphic evidence of how Olivia Hope and Ben Smart died. The other evidence was called mainly to place Watson with the pair and rule out anyone else. It might also be asked whether the fact that the two witnesses were called by the Crown bestowed an unwarranted degree of respectability.

Such questions lend support to those who want to ban the use of jailhouse witnesses. That course, however, would deny reality: some prisoners do talk to cellmates about the crimes they have committed. To discount such testimony would be to assume that every prisoner who presented incriminating evidence was lying. The Crown would certainly be criticised if it did not use such testimony and there was an acquittal.

Witness A, however, has seriously undermined the credibility of jailhouse witnesses in the eyes of the public - and future jurors. Perhaps only judges can ressurect their effective participation in the judicial process. Judges routinely rule on the admissibility of evidence. Should it not also fall to them, as a matter of course, to test the credibility of jailhouse witnesses and their evidence before it is presented to a jury? This extra test would not be foolproof. It would, however, allow jurors to again attach some credibility to such witnesses.