Here is the piece I wrote for the Chronicle of Higher Education on the American Psychological Association’s Independent Review on its collusion in the programme of enhanced interrogation. I focus here on the broader context. My book, Why Torture Doesn’t Work: The Neuroscience of Interrogation is available on Amazon. deals with these and related issues in depth. It will be released in November, 2015.
[Update: the piece below was chosen as a Science Seeker Editor Selection]
The Neuroscience of Interrogation: Why Torture Doesn’t Work
The American Psychological Association’s reported lapses are an ethical debacle, but a scientific one too.
By Shane O’Mara
The abuses of power and process at the American Psychological Association, reported in an independent review that was disclosed last week, are an ethical debacle. They are also a scientific debacle, because study after study shows that torture doesn’t work.
But if the disclosures are terrible, no one should find them unexpected. The futility of torture at eliciting intelligence in times of war and terror is a lesson civilization has learned and forgotten regularly. It is as if democracies live in silos, destined to make each other’s mistakes – from the French-Algerian conflict in the early 1960s, to the United Kingdom in the 1970’s, to the US in the 2000s. And each country learns the hard lesson it could have learned easily by conversing with other allies – but they don’t. In the hysteria of hostilities, the demand that “something must be done,” and never mind that the evidence, which shows from brain function to the annals of history, that torture is useless, inhuman, and degrading for the tortured and the torturer alike.
It’s a familiar story. A terrible attack or attacks occur against or within a democracy. People just going about their business — dining in a cafe, working at their office, riding a bus or train or plane — are grotesquely killed or injured. Then the frantic hunt for the perpetrators, the captures, imprisonment, and interrogations. Innocent and guilty alike are swept up in the dragnet. Inchoate dangers are presented, by those who should know better, of ticking-time bombs, mushroom clouds, the prospect of death and destruction all around. Volunteers — agents of government or shadowy contractors — emerge willing to torture. They are usually untutored amateurs who don’t know what to do, to whom, under what circumstances, supposedly to guarantee the loosening of tongues, to maximize information yield. They are untutored amateurs, because democracies do not maintain clandestine, specialised units with the requisite secret training, secret and involuntary human experimentation and all that institutionalised torture implies.The amateurs will be protected in what they do — for a while — by higher-ups. And eventually they will be dropped and ignored and prosecuted when it is all over — and it will, eventually, be over.
Gradually people adapt, because this is what they must do. Lives must go on and a certain level of security and aggressive retaliation becomes accepted. But slowly, as the panic fades, democracies start asking: “Is this us? Do we beat confessions out of people? Do we threaten the helpless and defenseless and their wives and children with mutilation and death? Do we extract information from people by force? Because something must be done?”
Then the recriminations start, and cases of torment and torture are examined in the cold light of day. The murdered and mutilated don’t get justice because the legal process against the captives has been so corrupted. Sometimes innocents have been jailed, sometimes not. And then the system starts to learn that innocent and guilty alike will confess to anything under torture. It learns that distinguishing the guilty from the innocent through torture is impossible, because both will confess to anything to get the extremes of stress and duress to cease. And the innocent and guilty alike will confabulate information of great quantity and worthless quality, because speaking at length causes the tortures to stop, at least for a time. Not the truth of what they say — just the fact that they say it, and say it at length.
Democracies, when they employ torture, deploy “white tortures” — techniques of torment and duress that leave no visible marks but which attack our integrated psychological, neural, and physiological functions at their core. The tortures employed are variations on fundamental themes, mixed and titrated according to taste: oxygen deprivation through near-drowning and suffocation; shackling and stress positions; extended sleep deprivation; freezing, cooling, and starving the body and brain; overloading the senses with loud noise and bright lights; a drip-feed assault on personal dignity through facial slaps and holds, enforced nakedness, and the imposition of adult diapers; the slow destruction of the integrity of personhood through social isolation, social deprivation, and a deliberate program of deindividualization; confinement in cramped boxes; threats involving guns and drills and attack dogs; pretended assaults on the loved ones of the captive.
And all the while, questions, questions, and more questions about the captive’s intentions, their confreres, ideology, bomb caches, and so on. The need for such information is the ethical rationale used to bolster the case for torture.
But it doesn’t work.
The core of torture as a theory is the idea that imposing severe stressors over some period of time results in loosening of the tongues of the unwilling. The captive has information which, by definition, is stored in the long-term memory system of his brain, and which can only be revealed through the medium of language. There is some (unspecified) conduction mechanism to transfer this information intact and in toto from brain to voice, facilitated by torture. However, extreme stressors force the brain away from the relatively narrow, adaptive range within which it operates. And memories are fluid, fragile, and subject to revisions and loss through time, fatigue, stress, and pain.
We know from multiple studies of combat and elite soldiers, certain patient groups, and normal populations that these stressors substantially compromise memory, mood, and cognitive function. Sleep deprivation, for example, is a most effective method for causing deficits in cognition. Volunteers undergoing sleep deprivation and combat soldiers enduring substantial sleep deprivation as part of their training, show large decrements in psychomotor and general cognitive function, as well as profound deficits in declarative memory. Torture profoundly disrupts the conduction mechanism transferring memories from brain to voice. Sleep deprivation also profoundly, negatively affects mood, which in turn compromises cognitive function further. Extended periods of sleep deprivation can cause polysensory hallucinations, psychotic-like episodes, and other neuropsychiatric phenomena. There is no evidence whatsoever that sleep deprivation in any way, shape, or form enhances access to memory about planning and intentions, or of personally-experienced events.
Other tortures use pain and the imminent threat of death. Titrating the dose of pain, so it is almost unbearable, but not so much that it causes the person to go into unresponsive shock, is very difficult. This is especially so for an amateur, who knows nothing of nociception — the nervous system’s processing of harmful stimuli — of c-fibers or synaptic transmission or pain gates in the spinal cord and thalamus. The brain will be overwhelmed by the processing of threat- and pain-related stimuli, caused in part by the massive and sustained release of a whole variety of neurochemical messengers and effects: endogenous opioids, neuropeptide Y, a panoply of stress hormones, all dampening reflective aspects of cognition and favoring immediate survival-related responses.
Studies of persons in severe chronic pain, and studies of the interaction between supervening states of pain, cognition, and memory demonstrate reliably that pain impairs cognition, memory, and mood. Predator threat through suffocation or near-drowning is similarly problematic. Oxygen restriction reliably draws activations away from brain regions concerned with higher cognitive function and memory in favor of brainstem regions concerned with reflexive responses supporting immediate survival — militating against detailed recall.
Chronic and severe stress compromises integrated psychological functioning, impairing recall, and facilitating the incorporation of leading questions, even if captive and interrogator both might not know this subtle process of incorporation has occurred. Chronic and extreme stress causes loss of tissue in the brain regions involved in memory (the hippocampal formation), and decreases activity in brain regions concerned with intention, planning, and the general regulation of complex behavior (the frontal lobes). Finally, it causes increases in the activity of brain regions concerned with fear and threat-related information (a core problem in post-traumatic stress disorder and generalized anxiety disorder). These changes by any reasonable standard constitute a form of organ failure.
And what of the torturers themselves? Because these events have happened in a democracy, there is no secret society of fellow torturers from whom to draw succor, social support, and reward. Sustaining physical and emotional assaults upon the defenseless and eliciting worthless confessions and dubious intelligence is a degrading, humiliating, and pointless experience. The torturers too become broken, and their fate is the bottle, pills, or worse.
Torture fails interrogation because torture is an assault on our core integrated, social, psychological, and neural functioning. Given what we know of the brain, memory, and cognition, it is little surprise that the yield from torture is so paltry. Torture corrupts and corrodes everyone and everything it touches: the guilty and the innocent, the perpetrators and the legal system. Experienced interrogators uniformly repudiate torture as a theory and as a practice, knowing that ethical and humane interrogations based on fostering respect yield usable, verifiable, and actionable intelligence.
The APA report is just the latest reminder of how willfully we can forget what we already know. However, it is also sign of our capacity to look at ourselves, and learn, however haltingly, from our mistakes.
Shane O’Mara is professor of experimental brain research at Trinity College, Dublin, and is a Wellcome Trust Senior Investigator. He is author of the forthcoming book Why Torture Doesn’t Work: The Neuroscience of Interrogation (Harvard University Press, November).