I wonder whether Alison Saunders, the director of public prosecutions, has heard this story.
It’s about Dorothy Martin, who was sure that on December 21 a flood would destroy the world. Aliens from the Planet Clarion told her it would. But she, and other true believers, would be rescued by a flying saucer. People gave up their jobs, sold their houses, joined Ms Martin and waited.
And then? Nothing. No saucer. Eventually the cultists realised that the moment had come and gone. The obvious conclusion was that the whole thing was nonsense. But that wasn’t how the cult reacted.
Instead they called the newspapers. Previously averse to publicity, they now proclaimed their message to everyone. They had received another message from Clarion. There wouldn’t be a flood because their devotion had saved the world.
There are two things worth knowing about this story. The first is that it was the basis for one of the most famous books in the field of social psychology When Prophecy Fails by Leon Festinger, Henry Riecken and Stanley Schachter. The second is that the December 21 in question was in 1954 and the book was published in 1956.
Advertisement
In other words, we have had more than 60 years to absorb the lessons from Festinger and his colleagues, yet it seems some are still struggling to do so.
Several prosecutions have recently collapsed at the last minute after it emerged that information in the possession of the police, but not disclosed to the defence, cast strong doubt on the complainant’s story.
There was the rape charge against Liam Allan, where police failed to disclose evidence that supported his story, and demonstrated that his accuser pestered him for “casual sex”. There was the case of Oliver Mears, who spent two years on bail accused of rape before evidence emerged that his accuser had lied.
And there was the case of the family doctor Stephen Glascoe, who faced a series of charges, including the alleged rape of a young girl, before they were dropped just before the trial opened. The accusations derived from so-called “recovered memories”. The accuser claimed the doctor had performed a forced abortion on her, although her accounts appear to have been based on watching Call the Midwife.
During the 18-month investigation into Dr Glascoe, police developed a close relationship with the woman involved, with one officer exchanging more than 1,000 texts and 500 emails with her.
Advertisement
The DPP has described these failures as “disappointing and irritating”, stressing the need to get the job done properly. She has added that she is confident that no innocent person has been jailed as the result of such an error.
That response isn’t good enough. I’m not suggesting that Mrs Saunders is unconcerned about people who have been unfairly accused. But has she learnt anything from academic research of the past 60 years into how human beings think?
Festinger’s work on the cult was the pioneering study in what is now a vast field. It suggests that once we develop a theory, it is very hard to shake. Indeed the stronger the counter-evidence, the harder we work to save our original idea. And the bigger the disaster brought about by being wrong, the harder we work to convince ourselves and others that we are right.
One piece of work on university fraternities showed that the more humiliating the initiation ritual for a fraternity, the more likely students were to value joining it. Who wants to think themselves the idiot who went through all that for nothing?
Understanding this is vital to gaining an insight in the behaviour of police and prosecutors. Police do not behave like Morse on television, piecing the puzzle together, before revealing the improbable killer right at the end. They develop a theory and then build a case.
Advertisement
Our “confirmation bias” — an elementary part of social psychology — explains how we seek comfort in every piece of evidence that confirms we are right and find a way of excluding anything that suggests we are wrong. Or even turning it around in our heads so that it becomes supportive. How does someone get it wrong after exchanging 1,000 texts with an unreliable complainant? Precisely because they’ve exchanged 1,000 texts with her.
As a result, police and prosecutors will not wish to disclose evidence that undermines their case and, crucially, may not even appreciate that it does. They may genuinely, but incorrectly, see the new material as irrelevant. That’s what confirmation bias does to you.
The failure to disclose crucial evidence to the defence is not just happening in a few cases. A report last summer from HM Inspectorate of Constabulary and HM Crown Prosecution Service Inspectorate suggests that the process is “routinely poor”.
So I was disappointed that Mrs Saunders suggested that no innocent person was in jail because of it. This would appear to add to the social psychology failure, a failure to think in statistical terms as most social science now does.
Only if the probability of wrongly convicting someone was virtually zero, and there was almost no randomness in results, can it be true that in a vast sample of convictions no one was wrongly imprisoned because of poor disclosure.
Advertisement
Indeed, Mrs Saunders comically misses the point when she suggests that anyone who feels they have been wrongly convicted because of disclosure failures should speak out. They are unlikely to know if there is vital evidence in their favour if they were never told about it.
Still, it’s not surprising that she sticks to the untenable theory that nobody has been wrongly jailed as a result of such failure. She is the DPP after all, so she wouldn’t want to accept an idea, however compelling, that prosecutions can be wrong. Festinger would understand her.
There needs to be a better internal system for challenging police officers and prosecutors as they become committed to a theory, and some external judgment about what material should be disclosed.
And, given that prosecutors seem to be relying on recovered memories despite all the work that Elizabeth Loftus has done to show how unreliable they are, it might be a good idea if everyone involved had some extra classes in social science.