Twenty-six years ago, 19-year-old Scots drama student Amanda Duffy was murdered after a night out with friends. In a landmark trial in 1992, there was only one suspect in the case, Francis Auld. Despite forensic evidence that could link Auld to Amanda’s death, he received a “not proven” verdict, which meant the jury deemed there was insufficient evidence to prove his guilt. Despite this acquittal, Amanda’s parents sued Auld in the civil court, and he was ordered to pay £50,000.
This third verdict, unique to Scots law, has long been controversial. High-profile cases such as the trial of Auld have led to many viewing it sceptically, with some suggesting the not proven verdict allows guilty individuals an extra chance of acquittal. These claims have put pressure on the Scottish government to consider abolishing the verdict, with Holyrood’s justice committee warning that the “bastard verdict” (as Sir Walter Scott called it) is on borrowed time.
But what are the effects of the not proven verdict on jurors? My research, the first study of its kind for ten years, aimed to establish just that.
The third way
The not proven verdict is to be found only in Scotland. Unlike the Anglo-American system, Scotland has three verdicts: guilty, not guilty and not proven. The confusion over the not proven verdict relates to the fact that Scottish law has not defined what the not proven verdict means, and it has the exact same legal outcome as the not guilty verdict, meaning the defendant is acquitted.
Besides being seen as confusing and providing an extra chance for acquittal, the verdict has also been said to undermine the presumption of innocence principle, which suggests that the Scottish three-verdict system breaks Article 6 of the European Convention on Human Rights.
Testing the verdict
My team wanted to test the claims that the not proven verdict may increase the chances of a defendant being acquitted. For this experiment we recruited 128 participants, and each participant acted as a juror in two separate mock murder trials.
In one trial, the jurors could give one of three verdicts: guilty, not guilty or not proven. In another, jurors could give one of two verdicts: guilty or not guilty. Participants were also asked to state the likelihood of the defendant’s guilt after each piece of evidence.
Our research found was that jurors were significantly less likely to give a not guilty verdict in the three-verdict system compared with the two-verdict system. There were also fewer not guilty verdicts than not proven verdicts in the three verdict system. Interestingly, there were no significant differences in relation to the number of guilty verdicts given across each of system; similar results have been found by researchers in the past. So if there wasn’t a significant decrease in guilty verdicts across the systems, then neither was there a significant increase in obtaining the acquittal verdict. The results showed that jurors did not give significantly different estimates of guilt across each system.
What this means
This research has a number of implications for not only the Scottish legal system, but also the Anglo-American legal system. First, the claims that the not proven verdict decreases guilty verdicts was shown to be unjustified. This would suggest that individuals who are guilty are not being given an extra chance of acquittal even when the not proven verdict is available to jurors. So we can claim that miscarriages of justice are not occurring because of Scotland’s not proven verdict, but are instead being caused by problems that plague the adversarial system all around the world, such as juror bias and lack of juror expertise in relation to forensic evidence.
Second, the fact that fewer not guilty verdicts were given in the three-verdict system suggests that the not proven verdict may aid juror communication. For instance, the not guilty verdict in the two-verdict system can be given in two circumstances: when the jurors think the person was probably innocent; and when the jurors believe the person is probably guilty but the evidence cannot support the claims against the defendant beyond reasonable doubt.
The not proven verdict reduces the ambiguity of the not guilty verdict, with this option in a three-verdict system meaning the jury actually believed the person to be innocent. This clarification of the interpretation of the not guilty verdict occurs because not proven may be interpreted as a verdict given when the defendant is presumed guilty but the jury is unable to choose this verdict because the evidence did not prove guilt beyond reasonable doubt. In other words, the not proven verdict can help jurors to communicate their belief of guilt to the judge more adequately.
Third, juror estimates of guilt were the same across the two different verdict systems, and these estimates were requested before jurors were given the opportunity to make a decision. It can, therefore, be said that the jurors presumed the same level of innocence about defendants regardless of how many verdicts were available, thus confirming that the not proven verdict does not undermine the presumption of innocence principle. This means that the Scottish three-verdict system is in line with Article 6 of the European Convention on human rights.
Based on these results – although more research is necessary, particularly in relation to crimes such as sexual assault – I would recommend that the Scottish government retain the not proven verdict. Other legal systems might do well to consider introducing the not proven verdict as it might aid jurors in communicating their belief of guilt more adequately than is currently available under the two-horse guilty/not guilty system.