What an unusual pleasure it is to find something important organised by the state that is working well - trials by jury. Opinion polls consistently show strong support for them. Now an extensive piece of groundbreaking research, which involved among other techniques the quantitative analysis of the outcomes of half a million charges against defendants in all Crown Courts in England and Wales, shows that the system passes every test.
That is a rare double: we value jury trials and they do their job. We revere Parliament, too, but we don't think it is particularly effective any longer. We strongly support the NHS but it lets us down a bit too often. We don't mind paying for the BBC, though we have our complaints. Probably only the monarchy is up there with jury trials, good in principle and good in practice.
Professor Cheryl Thomas's report, "Are Juries Fair?", commissioned by the Ministry of Justice, has a wealth of new information. It turns out that juries return guilty verdicts on almost two-thirds (64 per cent) of all charges. Of course it is hard to decide what the right proportion would be - not so high that juries give the impression of being too easily swayed by the prosecution's arguments, nor so low that they appear unduly lenient. But the figure quoted above becomes the more satisfactory the more it is analysed. The highest jury conviction rates are achieved where strong physical evidence is presented against the defendant as in cases of theft, drugs, falsification and deception.
It is much more difficult to secure convictions, for instance, in cases where the state of mind of either the victim or defendant is an issue. The offence of threatening to kill requires the jury to be sure of what the alleged victim plausibly thought was going on. In cases of attempted murder the jury must similarly be convinced that the defendant intended to kill someone, not just cause harm or injury. These are often difficult calls for juries, faced as they may be with conflicting testimony. So it is perfectly rational for the conviction rate to be lower.
The crucial test for the jury system is rape. It is widely assumed that the result is too often "not guilty". But this is incorrect. Juries find the defendant guilty in 54 per cent of cases, a higher rate of conviction than for other serious offences such as attempted murder, manslaughter, grievous bodily harm and threatening to kill.
But the question, "Are juries fair?" cannot be left just there. For these figures may disguise variations caused either by the ethnic composition of juries or the ethnic backgrounds of defendants or victims. This is obviously a difficult matter to appraise. To cut through to reality, Prof Thomas's team used simulations. Trials were filmed and edited so that only a specific factor (such as the race of the defendant) was altered in different versions of the case. Each version of the case was then shown to a large number of "juries" to decide. This is an artificial procedure but it is hard to imagine how else the team could have got near the truth.
Before coming to the results of the work on racial factors, there is a further consideration. …