A Consequentialist Argument against Torture Interrogation of Terrorists

 

Jean Maria Arrigo, Ph.D.

 

Joint Services Conference on Professional Ethics

January 30-31, 2003,  Springfield, Virginia

 

Abstract

      Much support for torture interrogation of terrorists has emerged in the public forum, largely based on the “ticking bomb” scenario.  I draw from the historical record, criminology, organizational theory, social psychology, and interviews with military professionals to envisage an official program of torture interrogation.  The quintessential element of program design is a sound causal model relating input to output.  I explore three, increasingly realistic models of how torture interrogation leads to truth: the animal instinct model, the cognitive failure model, and the data processing model.  These models expose the rational, mid-level social processes that lead from an official program of torture interrogation to breakdowns in key institutions—health care, biomedical research, police, judiciary, military, and government.  Stated most starkly, the damaging social consequences of a program of torture interrogation evolve from institutional dynamics that are independent of the original moral rationale.  Further, a legal, regulated program cannot eliminate use of rogue torture interrogation services, because they still serve to circumvent moral and procedural constraints on the official program. 

      In my view, it remains for consequentialist advocates of a U.S.  torture interrogation program to plan for social repair after severe and inevitable breakdowns of institutions and community.  The counterargument, of course, is that in a society destroyed by terrorism there will be nothing to repair.  That is why the actual causal mechanism of torture interrogation in curtailing terrorism must be elucidated rather than presumed.

 

Introduction

      Since the September 2001 terrorist attacks on the United States, much support for torture interrogation of terrorists has emerged in the public forum, largely based on the “ticking bomb” scenario.  National polls have reported 45%[1] and 32%[2] approval; a large web-site vote indicated 65% approval.[3] The appeal is that rare use of torture interrogation of key terrorists could thwart terrorist plans of mass destruction at minimal cost to civil liberties and democratic process.  Moreover, a strictly monitored legal program is expected to replace current, illegal covert programs.[4]

      Ethicist John Rawls proposed that consequentialist moral argument for a program of action should  include assessment of the practices required to implement that program.  For with careful attention to implementation, there is less danger of adopting means that do not actually reach the desired ends.[5]  As a social psychologist, this is the course I pursue.  I pass over foundational issues, such as the definition of “torture” and the morality of torture per se.  I do not reach as far as state-level issues, such as  international covenants banning torture.  Rather, I draw from criminology, organizational theory, the historical record, and my own interviews to explore the design, implementation, and consequences of such a program.  This strategy makes visible the mid-level social processes that lead from an official program of torture interrogation to serious dysfunctions in major institutions—health care and biomedical research, police and the judiciary, and the military and government.

      Policy studies demonstrate that the quintessential element of program design and implementation is a sound causal model relating input to output.[6]  Can we send only key terrorists into the torture chamber and produce at the other end only knowledge of terrorist plans and harmless prisoners?  For orderliness of exposition, I present three, increasingly realistic models of how torture interrogation leads to truth: (I) the animal instinct model, (II) the cognitive failure model, and (III) the data processing model.  The inadequacies of each model lead on to the next, with increasing involvement and compromise of key social institutions—that  is to say, with unintended inputs and unintended outputs of great consequence.  To these causal models, I append (IV) the rogue models of outsourcing torture interrogation to foreign or illegal information services, without regard for causal explanation. 

      In outline:

 

Practical Mid-Level Considerations for

Programs of Torture Interrogation of Terrorist Suspects

 

Causal Models of How Torture Leads to Truth   1.  Animal Instinct Model   In order to escape pain or death, the subject complies with the demands of the torturer.  The model fails when (a) physiological damage impairs ability to convey the truth and (b) torturers cannot control subjects’ interpretation of pain.	Prototypical Interrogation Scenarios   1. "Ticking bomb"   Subjects divulge their plans under sufficient torture.	Special Institutional Requirements   1.  Assistance of medical practitioners, before, during, and after torture sessions.
2. Cognitive Failure Model   he physiological and psycho-logical stress of torture renders the subject mentally incompetent to muster deception or to maintain his own interpretations of pain.  The model fails due to (a) time delays and (b) torturers’ inability to distin-guish truth from deceit or delirium.	2.  Fanatics, Martyrs, and Heroes   Subjects are resistant to coercion by means of pain or threat.	2A. Biomedical and psychological research into methods, detection, and concealment of torture.   2B. Establishment of a torture interrogation unit, with training of torturers in sophisticated methods.
3. Data Processing Model   Torture provokes ordinary subjects to yield data (both true and false) on an opportunistic basis, for comprehensive analysis across subjects.  The model fails when (a) analysts are overwhelmed by data and (b) torture motivates many new terrorists.	3. Dragnet Interrogations   Numerous weaker or less committed terrorists, associates, and innocents are detained and interrogated.	3A. Coordination of torture agencies with the police.   3B.  Coordination of of torture agencies with the judiciary.    3C.  Accommodation of torture units within the military and the government.

 

Non-Causal Models of How Torture Leads to Truth   4.  Rogue Models   Torture is emotionally, culturally, or historically inseparable from other methods, or it is one tactic among many in a hit-or-miss approach. The model fails when (a) biases and ulterior motives of torturers invalidate results or (b) torture tactics empower competing political or criminal entities.

Prototypical Interrogation Scenarios     4A.  Indiscriminately Brutal Foreign Intelligence Agencies   4B.  Covert U.S. Torture Interrogation Sites   4C. “Information Services” of Organized Crime.

Special Institutional Requirements     4.  Discreet government and military liaisons.

 

I.  The Animal Instinct Model of Truth Telling

      The January 2001 cover of The Atlantic Monthly asked in large print, "MUST WE TORTURE?"  Terrorism analyst Bruce Hoffman from Rand Corporation suggested we must.  He reported sympathetically on his meeting with a Sri Lankan intelligence officer—"Thomas"— whose terrorist opposition “eclipse[d] even bin Laden’s al Qaeda in professionalism, capability, and determination."  In one vignette,[7]

 

Thomas’s unit had apprehended three terrorists who, it suspected, had recently planted somewhere in the city a bomb that was then ticking away, the minutes counting down to catastrophe....  He asked them where the bomb was.  The terrorists—highly dedicated and steeled to resist interrogation—remained silent....So Thomas took his pistol from his gun belt, pointed it at the forehead of one of them, and shot him dead.  The other two, he said, talked immediately; the bomb, which had been placed in a crowded railway station and set to explode during the evening rush hour, was found and defused, and countless lives were saved.

 

      In this compelling ticking-bomb scenario, terrorists give up their secrets in a timely manner to escape immediate pain or death, a causal pattern I label the animal instinct model.  Ancient Greek courts placed great faith in testimony obtained through torture of slaves.  As Aristotle explained, free men can reason about the personal consequences of truth and false testimony, and they will testify, even under torture, to their long-term advantage; but slaves, deprived of reason, are compelled by their feelings to meet the torturer’s demands so as to terminate the pain.[8]  Harvard law professor Alan Dershowitz adds a moral element:  “...you want to cause the most excruciating, intense, immediate pain” but “with minimum lethality.”[9]

      The animal instinct model requires medical assistance at the torture site.  As explained by neurologist Lawrence Hinkle, who examined Korean War veterans after communist "brain-washing":[10]

     

      The human brain, the repository of the information that the interrogator seeks, functions optimally within the same narrow range of physical and chemical conditions that limit the functions of human organs in general....  Any circumstance that impairs the function of the brain potentially affects the ability to give information as well as the ability to withhold it.

     

In particular, the interrogation fails if the terrorist loses consciousness before revealing his plans.

 

Assistance of Health Professionals in Torture Interrogation

      Routine participation of medical personnel in state-sponsored torture interrogation has been documented worldwide.  Medical professionals determine the types of torture a person can endure, monitor the person for endurance under torture, resuscitate the person, treat the person to prepare for further torture, and administer non-therapeutic drugs.  To cover up torture, physicians falsify health certificates, autopsy reports, and death certificates.[11]  Studies of survivors show medical participation in the range of 20% to 40% of cases, or even "the majority".[12]  Coercive interrogation manuals, such as the 1983 edition of the U.S.  Army’s School of the Americas manual, call for the services of physicians and psychiatrists.[13]

      To obtain the physicians’ perspective, the Indian Medical Association surveyed a random sample of 4,000 members.[14]  Of the fifth who responded, 58% believed torture interrogation permissible; 71% had come across a case of probable torture; 18%  knew of health professionals who had participated in torture;16% had witnessed torture themselves; and 10% agreed that false medical and autopsy reports were sometimes justified. 

      Government-sponsored torture generates deep conflicts within medical communities and between governments and their medical communities.  In Turkey physicians have falsified evidence of torture in post-detention medical examinations, to the dismay of their patient-oriented colleagues.[15]  Prompted by the longstanding torture of Kurdish separatists, Turkish physicians have established five torture treatment centers.  Government authorities demanded patient lists and brought physicians to trial for criticizing the government.  Turkish physicians are harassed, arrested, and even tortured, as regularly reported in international medical journals.[16] 

      Proponents of a U.S. torture interrogation program must consider how to deal with both zealous cooperation and vigorous resistance from American health professionals, particularly those with access to sensitive information.  The challenges will be intense in some specialties, such as military medicine[17] and forensic medicine.[18]  In 2001, a group of Mexican forensic doctors confronted state-sponsored torture with a pledge to document torture “without distinction or discrimination of any kind, even under threat.”[19]  A recent opinion piece in the Western Journal of Medicine, by medical members of past U.S. delegations to the Organization for Security and Cooperation in Europe, observed:  "How sadly ironic it would be, now, for the United States to join the ranks of the countries it has long condemned for [torture]."[20]

      The larger lesson here is that a national security rationale for torture does not suspend, without tumult and schisms, the ethics codes of professions whose participation is required.  Judging from the experience of other countries, similar turmoil might be anticipated in clinical psychology, law, and journalism.[21]

     

II.  The Cognitive Failure Model of Truth Telling

      A greater obstacle to torture interrogation than bodily frailty of the subject is the mental resistance of fanatics, martyrs, and heroes to coercion.  A former U.S. intelligence liaison officer, Paul Copher, gave this example of an interrogation by the Turkish National Police [TNP] in the late seventies:[22]

     

      In 1979 the [TNP] team captured Nalan Gurtas.  She had personally killed 32 Turkish National Police officers.... Trying to obtain intelligence out of one of the most violent people on earth is a bit of trouble.... She screamed, spat, kicked, ripped up her holding facility, refused to talk....She tried to cut her wrists with a plastic knife; she ran full force with her head onto the wall.  She threw feces at all of us. Then the word came down to go to Level 1  [i.e., potentially fatal torture].  .... She stayed silent.... [During transport] she tried to jump from the speeding vehicle...after laughing she loved shooting police and Americans.  She was between the car and the ground when they fired 32 rounds into her, which was one hell of a hard shot to make.....

     

      Modern psychology maps the intractability of the psyche.  In general, people tend to react against those who constrain their freedom of action (reactance theory,[23] for instance, by substituting untraceable acts of sabotage.  People tend to become more dogmatic and tenacious in their belief systems when reminded of their mortality (terror management theory).[24]  And under extreme stress, as during a car crash, rape, or battlefield injury, people may dissociate, that is, experience the episode as if from a distance, temporarily protected from pain and awareness of consequences (traumatic stress theory).[25]  For  such reasons, Copher said he tried to interview terrorist suspects “before any heavy handed ex-Turkish farmer slapped them around.”[26]

      Even under the Nazis, torture interrogation failed to break dozens of high state officials and military commanders involved in late-war plots to assassinate Hitler.  According to Peter Hoffman’s History of the German Resistance: 1933-1945:[27]

 

Six months from the start of their investigations the Gestapo still had nothing like precise knowledge of the resistance movement.......This lack of information and knowledge is all the more astounding in that Himmler's men employed every means to extract confessions....  Moreover all forms of torture were used without hesitation....

 

 

Hoffman attributes the failure of the Gestapo to the “fortitude of their victims.”         

      Closer to home for Americans, during the months and years of North Vietnamese torture of American POWs, Commander James Stockdale estimated that under 5% of his 400 fellow American airmen succumbed to North Vietnamese demands for anti-American propaganda statements.[28]  These examples expose the bigotry in the expectation that key enemy terrorists will readily give up their plans and associates under torture.  

      The animal instinct model ultimately fails because the physiological experience of pain is mediated by individual and cultural interpretation.  Anthropologists identify the “punitive pain” envisioned by the torturer as only one of several competing interpretations.  In addition to the valor of “military pain;” exemplified by Nalan Gurtas, German resistors, and Stockdale, we recognize the “medical pain” of surgery, the “athletic pain” of the marathon; and the “sacrificial pain” of martyrs.[29]  A psychological study of 79 Palestinians detained by the Israelis identified seven meanings of their prison abuse, such as “struggle between strength and weakness” of character; “heroic fulfillment” of their role as liberators of their people; a “developmental task;” and “return to religion.” [30]

      In the cognitive failure model of truth telling, torture renders the subject cognitively incapable of maintaining his or her own interpretation of pain.  As understood by contemporary criminal psychology, overwhelming physical and mental stress create a state of disorientation in which the subject is unable to maintain a position of self-interest and becomes suggestible or compliant under interrogation.[31] Similarly, in the 15th to 18th Centuries, European courts employed juridical torture in the belief that bodily suffering eliminates self-will, which accorded with Catholic asceticism.  Absent self-will, the accused was presumed to expose the truth in involuntary testimony or nonverbal signs.[32]

      In contrast to the resistance of Stockwell’s airmen to North Vietnamese tortures, only 5% of the repatriated American  POWs held by the Chinese in the Korean War decisively resisted the cognitive disorientation—and reorientation— tactics of the Chinese; 15% “participated” to the extent they were court-martialed or dishonorably discharged.  The “brain-washing” regimen of the chinese included solitary confinement, lack of stimulus, unremitting observation by guards, sleep deprivation, ridicule, pressure to confess to unstated crimes, and interminable instruction in Marxism and Maoism, as well as deprivations and threats.  “Physical abuse and mistreatment were frequently the results of resistance, but seldom the prelude to participation,” “with only 3% of the “Participators” severely mistreated.[33]  Here I pass over a considerable literature pointing to the greater efficacy of noncoercive interrogation based on social skills: subtlety and finesse of interrogation,[34] sympathy with the subject,[35] appeal to the subject’s self interest,[36] and outright deceit and trickery.[37]

      The cognitive failure model demands cutting edge biomedical research into techniques of torture, to stay ahead of savvy terrorist organizations, and research into concealment and detection of torture, to stay ahead of human rights monitors.  Further, it demands a corps of trained torturers to apply the precision torture technology.  I explore these two institutional requirements next.

 

Biomedical Research for Torture Interrogation

      Twentieth Century researchers developed numerous techniques to induce disorientation:  psychiatric-pharmacological substances, electroshock, electromagnetic resonance, and others.[38]           But how could torture-interrogation research proceed in the United States, and what sort of scientists might participate?     

      Consider a Cold War strategy for a program of similarly stigmatized research, the CIA's mind control Project MKULTRA.  A 1977 Senate investigation into CIA research on human subjects disclosed that the CIA had simply contracted with  researchers at over 80 universities, hospitals, chemical and pharmaceutical companies, and research institutes through a front funding agency.  The Human Ecology Society, as it was called, was co-founded by the neurologist Hinkle who had examined the POWs “brainwashed” in Korea.[39]  One MKULTRA subproject sought "[s]ubstances which will enhance the ability of individuals to withstand privation, torture and coercion during interrogation...."  Another called for three research teams to compare various combinations of "straight interrogation, hypnosis, and drugs on subjects who had denied allegations known to be true."  Admiral Stansfield Turner, then Director of Central Intelligence, stated: “I believe we all owe a moral obligation to these researchers and institutions to protect them from any unjustified embarrassment or damage to their reputations which revelations of their identities might bring.[40]  The U.S. Supreme Court ruled that the identities should not be revealed for security reasons.[41]  The CIA research strategy therefore remained viable.

      The "war on terrorism" has produced a related research strategy.  The U.S. Defense Advanced Research Projects Agency (DARPA) hosted an invitation-only conference on March 11-13, 2002, called "Scientists Helping America." his According to a widely broadcast DARPA call for submissions, "The primary focus of the conference is to tap into new ideas of scientists and inventors...."  "Luminaries"  from industry and academia are invited to "come and interact with real special operation forces including US Navy SEALS and US Army Green Berets." [42]

      In a 1995 interview, I remarked to (now deceased) biological psychologist John Liebeskind on the potential contributions of his pain management research to torture interrogation.  In most fields of animal experimentation, anesthetics and analgesics are administered to mask the experience of injury; however, pain management researchers intentionally induce pain in laboratory animals to study the mechanisms of pain.  Liebeskind, a Fellow of the National Academy of Sciences and a great humanist, replied with candor and insight:[43]

 

I don’t think I am or would be in a good or well informed position to assess the ethical consequences of any military project I might be asked to help with.  They wouldn’t tell me the truth (and I certainly wouldn’t know if they had or not), and I wouldn’t really understand all the ramifications even if they did or thought they did.  So if we are developing something military and are in competition with other countries, I would feel an "us versus them" dilemma, and there surely aren’t very many "thems" I would trust more than us.

 

      Biomedical applications to torture interrogation may have a humanitarian theme.  A neuroscientist recently developed a “brain fingerprinting” technique.  Words or pictures relevant to an event are flashed onto a computer screen.  Measurement of the subject’s brain-wave responses to words or pictures flashed onto a computer screen determine whether the subject has knowledge of corresponding events.[44]  Copher suggested that we “use the brain fingerprint to find out who the terrorists are, then we torture them...,” thus avoiding torture of innocents—but, warned that, “like the polygraph, the machine would have a wide range of [desirable and undesirable] effects.”[45]

      Israel exhibits the motive for torture research to stay ahead of savvy enemies.  A former Palestinian prisoner proudly explained:  “We learned about all the types of ill-treatment and the techniques that the secret police use, and we learned to observe the behavior of enemy officers during our interrogation.”[46]  Turkey exhibits the motive for research into concealment and detection of torture to stay ahead of human rights monitors.  The European Union Commission postponed Turkey’s admission into the European Union because of the flow of torture victims from Turkey.  To gain the political and economic advantages of membership, Turkey attempted to develop torture techniques that leave no medical trace.  European forensic experts, meanwhile, employed successively more refined methods of detection, such as bone scintigraphy, ultrasonography, CAT scans, and MRI’s.[47]  Thus develops the race between techniques of torture and techniques of resistance and detection. 

      Torture spreads to foreign governments through exchange of torture technology.  Argentina gave classes in torture to Latin American military officers, and Brazilian torturers trained Chileans.[48]  The School of the Americas, with its coercive interrogation manuals, was widely known as a training school for Latin American dictators during the Cold War.  Amnesty International cites the United States as the largest international supplier of electroshock weapons—stun guns, electro-shock batons, and the like—to governments that practice electro-shock torture (for instance, $3 million worth to Saudi Arabia in 1990).[49]

      Another unintended consequence of biomedical research on torture is the opportunity for secret, illegal research on human subjects for other purposes.  Administrators of torture interrogation programs cannot prevent  alternative uses by their own biomedical scientists, who have negotiating power over valuable resources.[50]  Historically, government-sponsored ethics investigations of the CIA’s mind control Project MKULTRA and the Department of Energy’s radiation studies exposed extraneous, excessive, and criminal human subjects research.[51]

      Similarly, as will be seen later, torture interrogation is a resource that could not be denied to certain military, police, and government agencies for their own applications.

 

Establishment of a Torture Interrogation Unit

      Political scientist Harold William Rood, who for ten years taught field interrogation of prisoners of war in military intelligence schools, explained the institutionalization challenge as follows.  Many agencies must coordinate in a counterterrorist operation.  Reliability and accountability are therefore essential.  Outlaws and madmen cannot be hired as torturers by an otherwise orderly agency.  The torturers have to be well trained and professional:  "To deliberately torture a person over a period of six weeks is asking a lot of human beings. And therefore you have to recruit them very carefully."  This is the work of a highly trained team of interrogators.[52]  The Los Angeles Times  recently reported a former counter-terrorism official as saying that one of bin Laden’s top deputies, captured in March 2002, "will be put through the ringer for days and weeks....It will take at least 96 hours to get him to spill everything....If not, they will keep going...."[53]

      Training programs have been studied through interviews with former torturers in Greece,[54] Argentina, Brazil, Chile, and Uruguay,[55] Nicaragua,[56] and Israel,[57] to name a few.  As a composite picture, trainees are typically selected on the basis of their ability to endure hardship and pain, for correct political beliefs, trustworthiness, and obedience.  Often the young, the lower class, or the poorly educated are recruited, or even kidnapped.  Brutal training at the outset desensitizes trainees to their own pain, suffering, and humiliation.  Confinement and initiation rites isolate them from prior relationships.  Constant physical and psychological intimidation instill obedience.  Trainees first learn torture through social modeling in witness and guard roles.  They usually experience tension between their new roles and their previous values, and they variously resort to denial, psychological dissociation, alcohol, or drugs.  The efficacy of shame tactics tends to lead, as with the Israeli General Security Services treatment of Palestinians, to sexual tortures, which in turn contribute to stigmatization and corruption of torturers.  Training programs use dehumanization and scapegoating of victims to relieve the bad self-image experienced by many torturers.

      Criminologist Ronald Crelinsten, who studies the career paths of torturers, writes movingly of their travails in the closed world of the torture unit.  "The world of conventional morality and human decency....has been largely replaced by a new morality where disobedience"— including refusal to torture friends—"means punishment, disgrace, humiliation,...or even death."[58]  Alistair Horne, historian of the Battle of Algiers, concluded that torture "tends to demoralise the inflictor even more than his victim, and he gave harrowing examples of psychological impairment from the psychiatric record.  The French negotiator for the final peace settlement with Algeria, Louis Joxe, grieved:  "I shall never forget the young officers and soldiers whom I met who were absolutely appalled by what they had to do...."[59]  After the fall of the Pinochet regime in Chile, the navy and air force did not take back officers who had worked in the secret service but considered them to be “defiled.”[60]

      The utiliarian argument for torture interrogation, therefore, must justify the additional sacrifice of the torturers—made vulnerable to "perpetration-induced traumatic stress,"[61] severe social stigma, and possibly spiritual turmoil—or develop some hitherto-unknown transcendent regimen for training and after-care of torturers.  This sacrifice should be expanded to include also families, where secrecy, stress, and stigmatization take their toll over decades, and support staff—secretarial, custodial, and maintenance.  As Copher commented: “Although our team [of liaison officers] could become accustomed to the [Middle Eastern counterterrorist] interrogation process, the secretaries that received our hand-written analysis and interrogation observations were badly shocked.”[62]

      Rood further cautioned that it would be difficult to set up a program for torture interrogation without unintentionally incorporating an inside agent of the terrorists.[63]  Counterterrorist agencies depend on informants.  As in police detective units with official informant programs, there is the perennial risk that the informant will corrupt the agent and reverse roles, so that the informant becomes the handler and the agent becomes the paid informer.[64]  Recently, mafioso James "Whitey" Bulger,  unprosecuted murderer of 11 FBI informants, was found to have been running his own agents inside the FBI—including FBI Special Agent John Connolly, whose opposition to organized crime had raised him to eminence.[65] Considerations of futility are thus added to the sacrifice of personnel in a cost-benefits analysis.

 

III.  The Data Processing Model for Knowledge Acquisition

      The cognitive failure model for fanatics, heroes, and martyrs alerts us to the fact that torture interrogation provides only data, not actionable knowledge.  Commander James Stockdale ordered his fellow American POWs in Vietnam to  “resist to the point of permanent injury or loss of mental faculty, and then fall back on deceit and distortion.”  The standard of truth for intelligence agencies is not the one-shot revelation, which is always suspect, but convergent analyses of far-ranging data.

      The French General Paul Aussaresses, chief intelligence officer in the Battle for Algiers from 1955 to 1957, credits his large-scale, merciless, torture interrogation campaign with crushing the Algerian insurgency and stopping terrorist bombings.  Aussaresses’ name is often brought in to support the ticking bomb argument, but his memoir records cases of terrorists dying under torture with their secrets or exasperating him to the point of murdering them himself.[66]  He did not run a small-scale, precision, torture-interrogation program, as envisioned by advocates in the U.S.  He had an informant for each city block in the native section of Algiers.  Altogether, thirty to forty percent of the men were arrested and interrogated during the war, frequently under electrical shock to the sexual organs.[67]  Aussaresses ran a dragnet interrogation program, which swept up the peripherally involved and the uninvolved along with terrorist leaders. 

      Dragnet torture interrogation is the norm.  Soviet historians estimate that between 1936 and 1939 the Soviet government arrested 5% to 10% of the population, so as to suppress political opposition.  Under severe interrogations, with beatings, sleep deprivation, and isolation, almost everyone confessed, thereby providing public justification for the arrests.[68]  The widespread torture interrogation programs of Latin American dictatorships during the “dirty wars” against communism are well known.  Between 1987 and 1994, the Israeli General Security Service officially interrogated 23,000 Palestinians, of whom 94% were tortured or severely abused, according to a large interview study.[69]         

      Copher gave a feeling for this style of diffuse data collection on terrorists in Turkey a couple of decades ago.[70]

 

...We [American liaisons] were called when major terrorist arrests were made and we sat in.... We did not need to be present in the room but we were in the facility and monitored all of it.  Often we suggested that the person be pulled out of the session and brought to us.  We then started when it became clear the individual was a terrorist....

.     ..If [the locals told us] the subject went to “donkey heaven” [i.e., died under their interrogation], I would say, “It was already written by Allah,” and get on with the next truck load of subjects.  When a nation declares martial law, all the interrogators go into high gear and work until exhausted and then some....[Y]ou can conduct five or six normal-level interviews and suddenly you have a “holding pen” of several thousand people....

 

As a prototype to guide a torture interrogation program, the time scale of the ticking bomb scenario is extremely misleading.  In FBI experience, deterrence of  terrorist acts is a long-term affair, with informants, electronic surveillance networks, and undercover agents.  Operations must be tracked and allowed to play out almost to the last stage to comprehend their scope.[71]  The fanatics, martyrs, and heroes scenario errs, like the ticking bomb scenario, in its focus on key terrorists.  They are difficult to apprehend and likely to require great exertions from torturers.  Their numerous peripheral associates are much easier to apprehend and more susceptible to interrogation—whence the inevitable trend towards the dragnet interrogation model of knowledge acquisition.  Among the detainees will be many innocent or ignorant persons but these, too, are critical for comparison of nonterrorist with terrorist data.  The difficulty “from a purely intelligence point of view,” as noted by Horne, is that “more often than not the collating services are overwhelmed by a mountain of false information extorted from victims desperate to save themselves further agony.”[72]

      This is all to say that while torture of terrorist suspects indeed provides more data, it does not discriminate between truth and lies.  Turning to the related criminological studies, truth and lie detection falls into a conflicted area.  Contemporary police interrogation manuals provide time-tested schemas for truth and lie detection by eminent practitioners in the criminal justice system.  Manuals discriminate truth tellers from liars through numerous verbal and nonverbal behaviors, often linked to involuntary physiological responses.[73]  Partly contradicting this literature of interrogation expertise, experimental studies find that the short-term success rate of professional lie catchers—police officers, detectives, prison guards, customs officers, and the like—falls mostly in the 45% to 60% range, where 50% is the chance rate.  In these experiments, professional lie catchers only differ from nonprofessionals in their high level of confidence!  The professionals’ unwarranted confidence has been traced to lack of feedback in the workplace regarding the accuracy of their judgments and to their false reliance on behavioral signs of deceit, such as gaze aversion.  In fact, behavioral signs of deceit differ across liars and constitute only a small effect.  Accuracy of lie detection has been raised to the 70% to 85% level through arduous technical methods, such as film analysis of micro-facial expressions, content analysis of testimony, polygraphy by independent examiners, etc.  Yet even these methods can be counteracted by trained subjects[74] or may fail due to individual differences.Ï[75]  Torture, which exaggerates physiological variability in subjects, may complicate objective lie detection.

Because of the large number of detainees and the weak discrimination between the guilty and the innocent, dragnet interrogation requires much coordination between torture interrogation units and (a) the police, (b) the judiciary, and (c) the larger military institution and government.  I examine in turn these facets of an official torture interrogation program.

 

Coordination of a Torture Interrogation Program with the Police

      The U.S. "war on terrorism" has driven into close cooperation previously separated, or even estranged, police, FBI, CIA, and military intelligence agencies.  As of October 2002, the 18,000 police agencies, with 800,000 officers, vastly outnumbered the other agencies, for example, the 27,000 FBI agents, 58% of whom work abroad.[76]  Because of their numbers, networks, jurisdictions, and the urgency of action, in any torture interrogation program the police will be at the forefront of detention and interrogation of terrorist suspects.  Indeed, cross-national studies have found that expertise in torture typically passes from the police to military intelligence services.[77]

      What proportion of ignorant or innocent suspects are likely to be interrogated under torture?  Modern crime statistics indicate that among suspects arrested and charged with serious crimes, one-half to three-quarters are not convicted, depending on the state of jurisdiction.[78]  Under the proposed torture interrogation program, a detainee firmly believed to be involved in serious acts of terrorism will likely be tortured.  The secrecy and urgency of terrorist cases certainly cannot improve the rate of accuracy over serious criminal convictions, for the counterterrorist program rejects normal judicial safeguards—the right not to testify against oneself, the right to legal counsel, habeas corpus, bail setting, public hearings, and so on.  Moreover, in tracking terrorist operations it is customary to interrogate individuals just because they are acquainted with a person who has been detained, not because they are suspected of crimes.[79]  So an error rate of one-half to three-quarters of torture interrogees would be a low estimate.  For a long historical comparison, examination of court records for 625 cases of torture interrogation in France, from the 1500s through the mid-1700s, showed approximate rates of error—that is, no confession on the rack, under repeated drowning, crushing of joints, and the like—in 67% to 95% of cases, depending on the province.[80]

      One source of error is the nondescriptness of most terrorists.  In search of terrorist profiles, a 1999 government report on The Sociology and Psychology of Terrorism synthesized dozens of cross-cultural empirical studies.  Contrary to stereotypes, the report concluded that terrorists are “practically indistinguishable from normal people” in terms of outward appearance, they do not have “visibly detectable personality traits that would allow authorities to identify a terrorist,” they are not “diagnosably mentally disturbed”—except for an occasional psychopathic or sociopathic top leader.  Terrorist organizations  weed out the conspicuous and the mentally ill as liabilities.[81]

      How will the counterterrorist program uphold a monopoly on the use of torture?  Investigators of many other crimes—narcotics trafficking, serial murder, sabotage of information systems, espionage, financial scams....— will consider their own pursuits compelling, some at the level of national security.[82]  Both U.S. and British judiciaries have struggled for decades with the overwhelming ill consequences of coercive interrogation of suspects, which include false confessions and false testimony; police deception and manipulation of courts; failure of systems of oversight; and involvement of organized crime.[83]  The 1956 Miranda rights for suspects in the U.S. and the 1986 Police and Criminal Evidence Act in Britain[84] stipulate conditions of fair interrogation and (in principle) require judges to reject all but demonstrably voluntary confessions.[85]  On the whole, analyses of subsequent police interrogations have found negligible change in percentage of valid confessions.[86]  For reliability, professional police practice leans toward “extrinsic evidence independently secured through skillful investigation”[87] and away from confession evidence with its epistemic, ethical, and practical problems.[88] 

      A U.S. program of torture interrogation engenders the monumental task of discriminating between terrorist and nonterrorist criminal suspects, in spite of many overlapping criminal activities, and restricting torture interrogation to key terrorists.  Yet police departments are still struggling to transform a history of counterproductive coercive interrogation into a modern professional practice of efficient and accountable interrogation.

 

Coordination of a Torture Interrogation Program with the Judiciary

      Intelligence agencies around the world strongly resist interference from the judiciary and attorneys, who are often perceived as politically liberal, uninformed, infiltrated by the opposition, or, in any case, security risks for sensitive operations.  The Argentine General Acdel Vilas, who was active in counterinsurgency operations in the early 1970s, afterwards described how he circumvented judicial procedures.  He sent out plainclothesmen instead of uniformed officers to pick up suspects and passed only the insignificant suspects into the justice system.  The plainclothes approach, however, jeopardized his officers, and he had to cover for their deaths in irregular circumstances.[89]  Similarly, General Aussaresses wrote of an important terrorist who refused to confess:  “...I’m convinced he won’t say a word [under torture].  If there were to be a trial and he hasn’t confessed, he could actually walk away free, along with the entire FLN cadre.  So let me take care of him before he becomes a fugitive...”  Aussaresses himself hung the terrorist, as a weakly staged suicide.[90]  The Israeli Supreme Court found that interrogators from the General Security Service (GSS) routinely gave false testimony about coerced confessions to conceal its methods from terrorists, to secure convictions where evidence was lacking, and simply because false testimony succeeded.  GSS interrogators insisted that they only lied in cases of actual guilt, and, “[o]ne cannot clean sewers without dirtying oneself.”[91]  In the British criminological literature, this is called “noble cause corruption”:  self-sacrificing agents falsify cases and perjure themselves in court to convict people they believe to be criminals.[92]  Nevertheless, both British and Israeli security services have been shamed by notorious cases of the conviction of innocents after coerced confessions, false testimony by interrogators, and contrived evidence. [93]

       What about judicial monitoring of a torture interrogation program?  The U.S. judiciary, with its publicly elected and politically appointed judges, is not designed to hold its own against intelligence agencies, especially in matters of national security.  Dershowitz has proposed legalization of torture in ticking-bomb cases by requiring "torture warrants" judges, by analogy with search warrants.[94]  But the ticking bomb scenario itself does not afford judicial consideration of torture warrants for specific individuals on specific occasions.  A former Brazilian police officer explained: “It is necessary to get the information now because from now on to the future it might be too late.  And to save time, everything is valid.”[95]

      Within intelligence agencies, concealment of security operations is a professional virtue, not a moral flaw.[96]  A colleague’s obituary for Richard Helms, former Director of Central Intelligence, praised him “for standing up to the pressures” of a 1977 Senate investigation “to betray secrets that could be damaging to the country.”[97]  Intelligence professionals in the field are usually more skillful in methods of concealment than those dispatched to monitor them.  Copher provided an example:[98]

 

[E]very other year an inspector would stop by and ask what we all would do if ordered to eliminate a foreign national.  We would all say, “report them as per regulations.”  Then the inspector would leave.  Then what we were told to do was often in conflict with this idea.  In Lebanon we put our pro-U.S. warlord in charge and with the help of Shiites in Iran, [the foreign national] was assassinated.

 

      The consequentialist moral rationale for torture interrogation of terrorists provides no plausible mechanism for coordinating interrogators with the judiciary but would seem to support interrogators’ idealistic motivations to present false testimony in court and to thwart oversight.  A legal analysis of Israeli torture in the Occupied Territories concluded that in a liberal democracy torture subverts the rule of law and erodes other democratic ideals supported by the rule of law.[99]

 

Accommodation of a Torture Interrogation Program within the Military and the Government

      Rood reasoned from historical examples that, like the Nazi Deaths Head SS, an elite torture interrogation corps would be isolated from the regular military and intelligence.  Its commander would inevitably gain special powers, and his elite corps would have a destabilizing effect on the military and government.[100]

      Illustrative of destabilization of the military, under the leadership of several generals the Brazilian military gradually eliminated torture between 1975 and 1986 to save itself as an institution.  With counterterrorist        agencies working outside the law, torturers graded into “smugglers, blackmailers, and extortionists, and no one dare[d] to stop them.”  Torturers  scorned and controverted the chain of command, creating two factions and destabilizing the army.[101]  Illustrative of the destabilization of government, near the end of his life President Franklin Roosevelt commissioned a study that advised disbanding the Office of Strategic Services—the World War­II covert operations agency whose only rule was to win[102]—so as to preclude a “‘post-war Gestapo.’”[103]  Illustrative of the risk of excessive of powers, in 1962 the U.S. Joint Chiefs of Staff considered framing Cuba for a terrorist campaign on America, as a pretext for invading Cuba.[104] 

      Military wisdom cautions that the long-term potential of a weapon or tactic is more important than its initial purpose.[105]  Regardless of the initial purpose, resources invite new uses and are adopted by new users, as demonstrated by the spread of atomic weapons.  To this point, a recent issue of the Military Intelligence Professional Bulletin expressed concern that "The Patriot Act" approved by Congress on October 26, 2001, would "expand the definition of ‘terrorist’ to include non-violent protesters at an anti-war rally."[106]  Indeed, such a strategic expansion of the target group occurred under the Alien Enemy Act in World War II.  The incarceration of 7,000 German Americans partly resulted from a policy of arbitrary arrest to intimidate the German American community as a whole.  The deportation of 4,000 Latin Americans of German ancestry to U.S. internment camps partly resulted from their utility for prisoner exchange with the Germans and from local greed for their properties.[107]  The FBI and Bureau of Justice labeled people of German ancestry “potentially dangerous” on the basis of unverified accusations by neighbors, friends, business competitors,  disgruntled employees, and the like.[108]  Under the currently proposed torture interrogation program, such detainees would be in line for torture as well as incarceration.

      The larger institutional theme here is the difficulty of embedding a specially empowered domestic program of torture interrogation, with high secrecy, into a political system equilibrated by competing institutions.  In a national security framework, police, judicial, military, and government authorities are unprepared to check the excesses toward which the mandate and tools of a torture interrogation unit easily lead, according to both historical record and rational institutional processes.  Restated by a U.S.  Army chaplain, “We just cannot go there.  It would change our national identity.”[109]

 

IV.  Rogue Models of Truth Telling under Torture

      The grave, unintended consequences of overt state-sponsored programs of torture interrogation have made covert programs attractive.  In the rogue models, torture is not necessarily the causal factor in extracting the truth from captives but may be emotionally, culturally, or historically inseparable from customary tactics.  Or torture may be one tactic among many in a hit-or-miss approach permitted by lack of accountability for errors.

      One much discussed rogue option is outsourcing to foreign intelligence services.  In the Korean War, for instance, U.S. troops commonly turned over North Korean spies to South Korean intelligence for torture interrogation.[110]  The Association of Former [U.S.] Intelligence Officers acknowledged outsourcing in its October 21, 2001, newsletter:[111]

     

[S]ince Sept. 11, the CIA has arranged for 230 suspects in 40 countries around the globe to be jailed and questioned.  One notable aspect of putting possible terrorists in the hands of foreign security services is that [those] states ... use interrogation methods that include torture and threats to family members.

 

The loss of control over information and the creation of political debts discourages this option though.

      A second rogue option is the discreet use of illegal, government-sponsored torture.  Secrecy can be maintained through plausible deniability; isolation of torture sites; expendability of torturers; and initial recruitment of torturers trained elsewhere—abroad, in prisons, by abusive families.  A combat veteran, Kenneth Kendall, combining his “guinea pig” experience at the Nevada Test Site with his witness of torture interrogations by South Koreans,  conjectured the means of concealment of torture:[112]

 

Once you’ve got what information you need, then—it’s hard to do—but go ahead and finish killing them and erase them from existence, and then that would cover you from the dilemma of having to admit that you did something like this.  If there’s only a small number of people involved that could ever get out and say anything, then you can insist that they didn’t know what they were talking about.—“This guy sung his head off.  We didn’t have to do nothing with his wife and child.  We don’t know where they’re at.”  This is probably what I would advise, how this went and all.  Primarily based on my past experience.

 

      Numerous self-identified survivors of alleged government-sponsored torture—in such self-help organizations as the Advocacy Committee for Human Experiment Survivors-Mind Control (ACHES-MC), Stop Mind Control and Ritual Abuse Today (SMART), SurvivorShip, and Citizens against Human Rights Abuses(Cahra)—would agree with Kendall’s explanation.[113]

      The “information services” of domestic criminal enterprises offer a third rogue option.  It is a truism of criminology that organized crime supplies the goods and services for which there is strong demand but no legal avenue for supply.  Narcotics, “snuff” films, child pornography, transplant organs, artifacts from archeological sites, and illegal weapons are familiar examples.  In World War­II, the U.S. Navy turned to the Sicilian-American mafia for information about German U-boats off the Atlantic seaboard and the invasion of Sicily.[114]  Organized crime also runs “information services,” which include torture interrogation.[115]

      From an organizational perspective, these three rogue options are attractive because the services can be utilized according to need, with an elasticity difficult for state bureaucracies.  Moreover, training, monitoring, silencing, and disposal of participants can be managed with relatively little involvement or corruption of U.S. government institutions because few official liaisons need be involved.  The practical liability is the difficulty in breaking off or limiting relationships with rogue collaborators, on account of sensitive information held in “dirty hands.” In effect, terrorist information is bought through long-term tolerance of political and criminal wrongdoing, under the euphemism, “protection of sources and methods.”

      Consequentialists may argue that society is better off legalizing and regulating torture because it is impossible to prevent it.[116]  But an official U.S. program of torture interrogation could not eradicate rogue services.  They would still serve to circumvent ethical and procedural constraints on the official program.  And they are independently maintained by other political and criminal interests. 

 

Summary and Assessment of the Torture Interrogation Program

      Returning now to the proposed official program of torture interrogation, here is a rough schema of notable inputs and outputs, according to causal model.

 

Notable Inputs and Outputs of Programs of Torture Interrogation of Terrorists

According to Causal Model

 

I.  The animal instinct model of truth telling, for the ticking bomb scenario

Intended inputs • Terrorists with plans of mass destruction • Basic torture techniques • Ordinary police and military interrogators   Unintended inputs • Medical personnel at interrogation sites

Intended outputs • Fore-knowledge of plans of mass destruction • Harmless terrorist prisoners • Unharmed interrogators   Unintended outputs • Schisms and resistance in the medical community • Disabled and dead interrogees.

 

II.  The cognitive failure model of truth telling, for fanatics, martyrs, and heroes

Additional intended inputs • Terrorists resistant to coercion • State-of-the-art torture interrogation techniques • Teams of highly trained torturers   Additional unintended inputs • Biomedical research on torture techniques • Establishment of torture interrogation unit(s)

Additional intended outputs • General knowledge of terrorist operations and terrorist psychology   Additional unintended outputs • Inseparable mixtures of truth and lies regarding terrorist plans • Spread of technologies of torture to other nations and terrorist organizations • Opportunistic illegal, bio-medical research on human subjects. • Spread of technologies of torture, resistance, and detection to foreign governments • Ruination of torture personnel • Terrorist infiltration of program

 

III.  The data processing model of knowledge acquisition, for dragnet interrogations

Additional intended inputs • Terrorist suspects and possible associates • Bureaucratic apparatus for comprehensive data management and analysis. • Police assistance • Judicial oversight.   Additional unintended inputs • Falsely detained persons with no terrorist involvement or knowledge

Additional intended outputs • Understanding of social process of terrorist organizations, with actionable knowledge.   Additional unintended outputs • Tortured non-terrorist prisoners and corpses. • Opportunistic torture interrogation of other classes of prisoners and political dissidents. • Institutional degradation of the police. • Institutional degradation of the judiciary. • Institutional degradation of the military and government.

 

      Policy studies demonstrate that program success in balancing intended outputs against unintended outputs depends on formal access by outsiders and on provisions for independent evaluation studies.[117]  A  criminological study of police interrogation methods over a 50-year period concluded that “all police interrogations should be witnessed by observers who are both independent of police and empowered to act on behalf of suspects by preventing escalation of inherently coercive pressures.”[118]  In the case of a torture interrogation program,  such a safeguard is impossible for security reasons. 

      In my view, it remains for consequentialist advocates of a practical U.S. interrogation program to plan for social repair after the inevitable breakdowns of institutions and community.  In country after country where national security threats drove torture of domestic enemies —South Africa, Chile, Argentina, Uruguay, El Salvador, Guatemala, Ireland  ...—human rights researchers have shown the inadequacies of various attempts at social repair—the criminal trials, truth commissions, reparations to victims, community mourning rituals, ....  The great majority of bystanders constitute an intractable element.  Their passivity facilitated the social breakdown but they cannot be held accountable for the repair.[119]  Another intractable element is the disparity between repair policies effective at the state level and policies effective at the individual level.  A victim group accused the South African Truth and Reconciliation Commission of “‘reconciliation showbiz’,” for substituting disclosure of wrongdoing for prosecution of  wrongdoers.[120]  The National Association of Radiation Survivors accused the 1993 U.S. President’s Advisory Committee on Human Radiation Experiments of  “giving the government what the government paid them for”—a cover-up job.[121]  When grievances, such as torture of nonterrorists, cannot be remedied by the state, scapegoating, cover-up, discrediting of victims, and token reparations easily emerge as fraudulent social repair tactics.

      The counterargument, of course, is that in a society destroyed by terrorism there will be nothing to repair.  That is why the actual causal mechanism of torture interrogation in curtailing terrorists must be elucidated rather than presumed.  A criminological analysis of 500 British court cases found that police interrogation of defendants contributed little to discovery and conviction.  Rather, the study concluded that interrogation fulfilled certain psychological and administrative needs and that “police perceptions of reality dominate the criminal process.”[122]  Unless the question of efficacy of torture interrogation of terrorists is settled, advocates are liable to the charges of making excuses for inflicting suffering on others[123] and imposing torture for social control.[124]

      The historical record is not entirely clear but suggests that initial gains from torture interrogation are later lost through mobilization of moral opposition, both domestically and internationally, and through demoralization or corruption of the torturers and their constituencies.  In a final assessment of the efficacy of the French torture campaign in Algiers, Horne wrote:  "the stunning, cumulative impact it had was materially to help persuade public opinion years later that France had to wash her hands of the sale guerre."  Winning the Battle of Algiers meant losing the war.[125]

 

Conclusion

      Alan Dershowitz closes his essay on “Torture of Terrorists” with this challenge:[126]

 

Had law enforcement officials arrested terrorists boarding one of the airplanes [of the September 11 disaster] and learned that other planes, then airborne, were headed toward unknown occupied buildings, there would have been an understandable incentive to torture those terrorists to learn the identity of the buildings and evacuate them.

 

I would like to close my essay with a reply. 

The individual law enforcement officials, of course, can make their own moral choices and take their own risks.  But if it is state policy to torture the terrorist, then the policy should be rational and the torture interrogation proceed with a reasonable chance of success. 

      Terrorists selected for such a role—like most American POWs in North Vietnam—can probably stand up to commonplace tortures from untrained staff for a long time.  The use of sophisticated techniques by a trained staff entails the problematic institutional arrangements I have laid out:  physician assistance; cutting edge, secret biomedical research for torture techniques unknown to the terrorist organization and tailored to the individual captive for swift effect; well trained torturers, quickly accessible at major locations; pre-arranged permission from the courts because of the urgency; rejection of independent monitoring due to security issues; and so on.  These institutional arrangements will have to be in place, with all their unintended and accumulating consequences.  Then the terrorists themselves must be detected while letting pass without torture a thousand other criminal suspects or dissidents, that is, avoiding a dragnet interrogation policy. 

      The moral error in reasoning from in the ticking bomb scenario arises from weighing the harm to the guilty terrorist against the harm to the prospective innocent victims.  Instead, the harm to innocent terrorist victims should be weighed against the breakdown of key social institutions and the state-sponsored torture of many innocents.  Stated most starkly, the damaging social consequences of a program of torture interrogation evolve from institutional dynamics that are independent of the original moral rationale.

 

Acknowledgments

      For conversations, correspondence, and interviews with former and active military professionals, I thank:  William Casebeer, Barak Cohen, Paul Copher, Ernest Garcia, Kenneth Kendall, Tony Pfaff, and Harold William Rood.

      For responses to my query on the intelligence list-serves of Cass Publishers’ Intelforum (http://www.intelforum.org/) and the Association of Former Intelligence Officers  (http://www.afio.com), I thank:  Robin Bhatty, Drake Currie, Mike Dravis (moderator), Harlan Girard, Stephen Godjas,  Brendan Howley, Howell McConnell, Bill Oliver, Pauletta Otis, Lewis Regenstein, Richard Thieme, Allen Thomson, Douglas Vaughn, Nigel West, and Mike Yared.

      For responses to my query on the list-serve Ask-a-Philosopher, sponsored by Sheffield University for the International Society of Philosophers (http://go.to.ask-a-philosopher‚ I thank:  Sean Allen-Hermanson, Steven Bullock, Anthony Flood, Hubertus Fremery, Geoffrey Klempner (moderator), Derel Dov Lerner, Harry Lesser, Gerald Marsh, Matthew del Nevo, Michael Neumann, and Duncan Richter. 

 

For review of the manuscript, I thank John Crigler.

 

 

NOTES