Torture and Cognitive Dissonance

When my information changes, I alter my conclusions. What do you do, sir?

—JM Keynes

A couple of recent, unrelated articles caught my eye. The New Scientist carried a review of ‘Why Torture doesn’t work’, a forthcoming book by Shane O’Mara, an academic at Trinity, Dublin. The scientific answer is that torture doesn’t work, in the sense of getting at ‘truth’, for people will say anything in an attempt to end their torment. People will falsely confess, something that’s “alarmingly easy to produce”. Torture is euphemistically referred to as ‘enhanced interrogation’ by the US federal authorities. The police also use methods of ‘enhanced’ interrogation.

The second article was in the New Statesman (here) and was about cognitive dissonance, the mental tension resulting from the mismatch between new evidence and a pre-constructed ‘reality’. The article describes a convicted rapist and murderer. Years later, DNA testing showed that his DNA didn’t match what was found at the crime scene. At an appeal, the prosecutors introduced the presence of an hitherto unknown and unidentified further man who had raped the child, who the convict then came across, raped using a condom (thus leaving no DNA) and murdered. This remarkable theory has become so common that’s it now known to defence lawyers as the “unindicted co-ejaculator”. It’s also very apparent that the authorities are reluctant to confirm that DNA has been recovered from the scene, and reluctant to permit testing. Even today, one of the original investigators still believes in the guilt of the falsely convicted man. (The article also notes the calls for an ‘innocence commission’; and the UK’s Criminal Cases Review Commission and the ‘Birmingham Six’ and the ‘Guildford Four’; in both of which coercive methods of interrogation had been used.) Despite new evidence, new facts, investigators and prosecutors can be very reluctant to discard their theories.

The trial of Stephen Ward (and here) involved the coercive interrogation of witnesses such as Mandy Rice-Davies and Christine Keeler; the police were ordered to find evidence of criminality on Ward’s part—rather than the usual way of policing then, finding evidence of crime and looking for a perpetrator, for Stephen Ward was the sort of unreliable person who might damage or unseat the government.

Surely not in Ireland? I wrote previously about the murder of Patricia Curran and how an entirely innocent man was convicted of her murder. This man had been intensively interrogated by Chief Superintendent Capstick who was convinced of his guilt and who extracted a ‘confession’.

The tragedy that is the ‘Kerry Babies’ is more recent; the surrounding investigation was later the subject of a Tribunal (in the UK, an Inquiry). A dead baby was found on a Kerry beach; he had been stabbed multiple times. The Gardaí trawled the county looking for the mother; a young, unmarried woman, Joanne Hayes, became a suspect. She had previously given birth to a child whose father, Jeremiah (Jeb) Locke, was a married man, and though she had apparently been pregnant, there was no sight of a second child. She and her immediate family were interrogated and confessed to the murder of the baby boy. They retracted their confessions soon after. In a confused series of events, the young woman also described delivering a baby at her home, concealing the birth.

The Gardaí then discovered the body of this second baby at the young woman’s home. They seem to have decided that the woman had given birth to twins, and disposed of both of the infants. However, the baby found on the beach had Landsteiner blood group A; both the Joanne and Jeb, who didn’t deny paternity, were blood group O. In other words, the baby on the beach wasn’t the offspring of the Jeb Locke and the Joanne Hayes. Well, you might think so.

You might also have thought that the simple explanation of this was that two women had given birth, one having concealed the baby at home, and the other, alone or with an accomplice, had murdered the boy found on the beach. The Gardaí thought otherwise; Joanne Hayes had indeed given birth to twins, but these had been fathered by two different men. This is known asheteropaternal superfecundation. Superfecundation, where separate acts of intercourse give rise to multiple, fertilised ova is well known in animals. It’s more difficult to be certain in humans where multiple births are much less common. It can be seen in fraternal, non-identical or dizygotic twins where two different ova have been fertilised; identical, monozygotic twins develop from one fertilised ovum. Perhaps 2% of non-identical twins are the result of superfecundation; twins occur in roughly 1.25% of pregnancies. There were just over 64,000 live births recorded in Ireland in 1984; thus about 400 sets of twins (i.e. 800 babies), either monozygotic or dizygotic; at a maximum, 10 babies could have been a result of superfecundation. Which is all a long-winded way of saying that yes, superfecundation occurs, but is rare. The Tribunal also wondered if blood groups could change after immersion in the sea, in what seems like a desperate attempt to confirm the superfecundation theory. It was only late in the day that an expert indicated that as Joanne did not have anti-A antibodies in her blood, she could not have been the mother of the baby on the beach.

Nell McCafferty has written a feminist account of what happened here. She describes how the Tribunal was composed entirely of men; and how one of these, a married man, during the Tribunal proceedings took to his bed in a Tralee hotel a woman who wasn’t his wife. DNA testing wasn’t available in 1984; Joanne Hayes has subsequently called for it, to confirm her innocence, as has one of the original investigating officers (here), for he still believes that the superfecundation theory is correct. To date, DNA testing hasn’t been done.

Which brings me back to where I started; coercive interrogation of people doesn’t work; people will say anything at all so that their ordeal will end. They might well say what is true, but equally they will agree with any suggestion, however outrageous. Confessions extracted under such conditions must be regarded as very suspect; but then, whose word do you take for the ‘conditions’? The word of the torturers or their victims?

Likewise, when the authorities are convinced of the correctness of their suspicions, what does it take to change this? New and better evidence, no matter how persuasive, may well have no effect. Once we know something to be true, it seems that no amount of ‘evidence’ will cause us to deflect from our beliefs. Do the authorities still condone the use of “enhanced interrogation”? Or do they also have cognitive dissonance?