Metin Basoglu's blog on war, torture, and natural disasters

Waterboarding is severe torture: Research findings

In my previous post I reviewed the misconceptions that characterize the debate on torture renewed by the movie Zero Dark Thirty and presented some research evidence showing that waterboarding is one of the most physically and mentally distressing forms of torture. In this article I expand on the issue of waterboarding and review further evidence that has not been previously published. To add some touch of ‘human interest’ to this post I will also tell the story behind the research that generated this evidence and why such important evidence remained unpublished to date.

Until the early 1990s there were few methodologically sound studies of torture. In 1993 we conducted the first controlled study to investigate the mental health effects of torture. The study, conducted in Turkey, compared 55 tortured political activists with 55 non-tortured political activists. The results (published in the American Journal of Psychiatry in 1994) demonstrated an association between torture and posttraumatic stress disorder (PTSD).

In this study we used an Exposure to Torture Scale to obtain information on 45 different form of torture most commonly reported by survivors. As the best way to understand the impact of various forms of torture is simply to ask people who have experienced them, we asked the study participants to rate the intensity of distress they experienced during each torture event they reported using a 0-4 scale (0 = no distress at all, 4 = extreme distress). These ratings allowed us to examine which forms of torture were most distressing and conducive to PTSD.

The analyses revealed quite remarkable findings. Most strikingly, among all 45 different forms of torture, torture involving suffocation or asphyxiation was the strongest predictor of PTSD. Furthermore, repeated exposures to asphyxiation had a cumulative effect on PTSD symptoms (which meant the more exposures the survivors had to asphyxiation, the more severe their PTSD symptoms were). The study participants had been exposed to asphyxiation torture on average 9 times (ranging from 1 to 11).

We wrote up these results for publication in 1995. Below are some paragraphs from the discussion section of the manuscript.

Perceived distress during the following forms of torture was related to subsequent post-traumatic stress responses: torture involving threat to life (e.g. asphyxiation, verbal threats of death), torture manipulations designed to maximize loss of control and helplessness (e.g. restriction of movement, beating, and exposure to vermin infested surroundings, with the ultimate consequence of the latter being lack of control over exposure to sources of infection and disease), humiliation (e.g. verbal abuse, prevention of personal hygiene, and prevention of urination/defecation with the ultimate consequence of the latter being wetting or soiling oneself), and violation of personal beliefs or values (e.g. exposure to loud music).

Of all forms of torture, asphyxiation was the strongest predictor of PTSD symptoms. This is consistent with observations of fear of suffocation being an extremely intense and common fear in normal subjects. It has also been suggested that fear of suffocation is an evolved alarm system triggered by increasing PCO2 and brain lactate values. That repeated exposures to asphyxiation was associated with more re-experiencing symptoms might be seen as consistent with the notion that repeated trauma which triggers a fear of suffocation may have a sensitizing effect by lowering the reaction threshold. Given Klein’s emphasis on the primacy of the fear of suffocation for triggering an evolved alarm mechanism, it may not be too surprising that repeated exposure to only two traumatic stressors, asphyxiation and threat of rape,  was related to the re-experiencing symptoms of PTSD.

One might wonder why several of the most distressing forms of torture were not predictive of PTSD. Release of endogenous opioids is a well-known physiological defense against extreme pain and, with extensive exposure to uncontrollable physical stressors, the opioid release becomes conditioned to the onset of painful stimulation so that it occurs very rapidly following their onset. This phenomenon may have protected the survivors against the traumatic effects of three of the most excruciatingly painful forms of torture: being hanged by the wrists (often with hands tied at the back – so-called Palestinian hanging), electrical torture, and beating of the soles of the feet (falaqa). This hypothesis is consistent with self-reports of torture survivors who describe a general numbing feeling in the body during electrical torture.

It was also noteworthy that rope bondage was associated with the highest distress, and that restriction of movement was associated with PTSD symptoms. Indeed, some survivors report that the effects of painful stimuli are maximized when the body is immobilized. This is consistent with observations which suggest that restraint in animals potentiates the effects of exposure to uncontrollable stressors.

Humiliating treatment and attacks on personal integrity and morals were related to PTSD symptoms. Prevention of personal hygiene often meant depriving the female detainees of hygienic care during menstruation, which possibly explains in part why this event category was perceived as more distressing by women than men. Humiliation may induce feelings of helplessness in the individual through not being able to act on anger and hostility generated by such aversive treatment. There is evidence that animals and humans respond with anger, hostility and aggression to threats to physical and psychological well-being. Furthermore, the ability to aggress during uncontrollable stress can dramatically reduce the impact of the stressor in animals. This idea is supported by anecdotal reports of some torture survivors, which suggest that expression of anger and hostility towards the torturers alleviates distress during torture.

Among all torture events only three (asphyxiation, sleep deprivation, and threats of rape) showed a cumulative traumatic effect…The cumulative effect of repeated exposures to asphyxiation is possibly due to the highly aversive physiological experience of suffocation. Fear of dying combined with the actual sensation of suffocation appears to be more traumatizing than appraisal of threat to life on solely a cognitive level.

Finally, the present study highlights the importance of gaining insights into how torturers operate. As international pressure on torturers grows, more and more sophisticated methods of torture, mainly of a psychological nature, are being developed to avoid leaving physical scars on the tortured individuals. The parallels between experimental paradigms of anxiety, fear, and learned helplessness in animals and the way torturers operate in inducing helplessness in the victim, are indeed striking. Such sophisticated knowledge possessed by torturers may in part be a legacy of centuries long practice of torture, passed on from generation to generation, but it may also reflect systematic training of torturers. We hope that studies of this kind will promote public awareness of this serious issue.

The manuscript – written years before 9/11 and the current debate on torture – was never submitted for publication, because the local human rights organization that supported the project by referring cases vehemently opposed the idea in case the findings – particularly those concerning asphyxiation torture – informed torturers in designing more effective torture. My position on this issue was that torturers already knew a great deal about torture and that the findings simply reflected torturers’ existing knowledge and experience. I also argued that, as behavioral scientists, we had an obligation to make the study public, so that the findings could be used in human rights efforts against torture. The findings also had important implications for effective treatment of survivors. In an effort to resolve this disagreement, a meeting was held, also attended by about 30 study participants. After a heated debate, the study participants were also divided among themselves, some arguing against publication, while others questioning why the study was conducted in the first place if it was not going to be published. As no unanimous decision could be reached, we had no choice but to refrain from publishing the manuscript.

In 1997 we launched a new project in former Yugoslavia countries to investigate the mental health impacts of war and torture trauma. The study (main report published in the Journal of American Medical Association in 2005) involved 1,358 war survivors, 279 of whom had captivity and torture experience. Thus, we had another opportunity to examine certain important issues concerning torture, as I will discuss later.

Then came 9/11 and the so-called war on terror. The US government adopted a narrow definition of torture and the debate on waterboarding began. In 2006 I received an email from Dr. Robert Banzett, a Harvard Medical School scholar who conducts research on dyspnea (shortness of breath) caused by medical conditions. Some years earlier I had sent him a copy of the unpublished manuscript.

Like most Americans I have had concerns about the use of torture by people in our employ, and due to my research focus, I have special concern that the severity of suffocation torture is underestimated in the public discussion… The public, and politicians, have a clearer concept of the inhumanity of painful torture, and, as shown by the overwhelming vote in congress to ban it, I think they find it abhorrent.  If they knew how suffocation torture felt, and what its consequences are, I believe they would also abhor it.  We have some data on the ‘affective dimension’ of air hunger…Clearly, even under controlled laboratory conditions it is a very uncomfortable and threatening experience.

Apart from our own studies on what air hunger feels like in laboratory subjects and patients, one of the key pieces of information is the manuscript you sent me in ’99 entitled “Perceived distress during torture and its relationship to post traumatic stress responses”.  As far as I can tell, that study has not been published.  (I know that one of your concerns was that the information would get into the hands of the wrong people, but apparently the wrong people already know it. Maybe if the right people knew it, they would help put a stop to it.) Have you published anything else that shows evidence of the trauma inflicted by suffocation torture, or do you know of others who have?  Do I have your permission to show your [manuscript] to Congressman Frank? 

Robert B. Banzett PhD, Associate Professor

Physiology Program, Harvard School of Public Health

Beth Israel Deaconess Medical Center, Harvard Medical School

Dr. Banzett was absolutely right! It was clear to everyone by that time that the “wrong people” already knew it and there was no point in withholding the results any further. By that time we were also about to publish new data on tortured war survivors showing asphyxiation at the top of the list of the most distressing and uncontrollable forms of torture (see previous post). So I agreed to the dissemination of the findings as necessary. Sometime later Dr. Banzett wrote back:

We sent the message below to Barney Frank, Raul Grijalva, Edward Kennedy, John Kerry, John Kyl, and John McCain (our Senators & Representatives)

In following the recent confirmation hearings for Attorney General, we have been concerned with the lack of clarity about “waterboarding”, and suffocation procedures in general. We write to you as experts on the sensations experienced by humans deprived of sufficient air.  Our goal is that there be a clear understanding of the sensations experienced to aid informed judgment as to whether forms of near-suffocation, such as “water boarding” constitute torture.

 For 20 years our laboratory has studied the symptom known to medical professionals as “dyspnea”, colloquially “shortness of breath”.  Dyspnea is the leading symptom of cardiopulmonary disease.  One of the underlying sensations of dyspnea is “air hunger” – the sense of needing more air – in its extreme form it is the sense of suffocation, and is one of the primal biological drives (Denton 2006).  We are able to reproduce this sensation under safe conditions in the laboratory, with the important ethical caveat that the subject is always allowed to immediately terminate the experience whenever it becomes unbearable. 

 An important element of torture, of course, is that the victim has no control.  Data from our lab show that subjects liken the sensation to “being about to die”, and comment that if they were not in total control with total confidence that they were safe, it would be a horrible experience.  In fact, Senator McCain, who has personal experience in these matters says of “waterboarding,” that it is a mock execution and “The memory of an execution will haunt someone for a very long time and damage his or her psyche in ways that may never heal. In my view, to make someone believe that you are killing him by drowning is no different than holding a pistol to his head and firing a blank. I believe that it is torture, very exquisite torture.”  There are actual data from torture victim interviews that show the persistent psychic damage of suffocation torture is in fact worse than mock execution with a pistol (Basoglu and Mineka 1999).  The reason for this is that the concept of execution with a gun is intellectual, and after a few tries the victim learns that she or he is unlikely to actually be killed.  Air hunger, on the other hand is a primal sensation even more basic than pain (Denton 2006), and is “hardwired” into primal sensory areas of the brain (Banzett et al. 2000).

 Scientifically, there need be no doubt that suffocation techniques meet the current [Bush] Administration’s own definition of torture.  The purpose of this letter is to bring our scientific expertise to bear specifically on the issue of suffocation torture.  We are prepared to speak with any or all of you to clarify and expand this brief explanation, or to provide the references upon which is based.

Yours Truly,

Robert B. Banzett PhD, Associate Professor of Medicine, Harvard Medical School

Robert W Lansing, Professor of Psychology Emeritus, University of Arizona


Banzett, R. B., H. E. Mulnier, et al. (2000). “Breathlessness in humans activates insular cortex.” Neuroreport 11(10): 2117-20.

Basoglu, M. and S. Mineka (1999). “Perceived distress during torture and its relationship to post-traumatic stress responses.” Personal Communication, unpublished work not submitted because of ethical concerns of authors, available for inspection.

Denton, D. (2006). The Primordial Emotions: The Dawning of Consciousness. USA, Oxford University Press.

When I decided to write this post I emailed my American colleagues asking for their permission to quote this email correspondence. Until then, I did not know anything about the response to their letter. In his reply, Dr. Banzett noted: “We never…got any reply to our letter from the senators — interesting and perhaps ominous.”

In a series of recent email exchanges discussing the reasons why waterboarding is so intensely distressing, Dr. Lansing wrote:

The complexity of torture (physical and psychological) makes me realize how simplistic most public discussions of this are, especially when mixed with political motives. This makes your work important for the education of everyone with an interest in this problem. One potential source of misunderstanding may be the general notion of that the agony of asphyxiation of suffocation is simply the lack of oxygen, indeed many of our subjects in the air hunger experiments describe their feeling as “not getting enough oxygen”.  Our understanding that air hunger results when the degree of lung inflation does not meet metabolic demands (whether due to hypoxia or increased CO2) emphasizes the role of limited breathing. In addition to the air hunger that is induced, depriving the person of the natural response to breathe more must produce a profound sense of loss of control in addition to the loss of behavioral control (the ability to change or leave the situation). We saw both of these effects in one of our subjects, an experienced pulmonary physician with full knowledge of our experiment and completely trustful of the experimenters. When he felt he wasn’t getting sufficient ventilation his behavioral response was to grab the ambu bag and try to ventilate himself. The torture victim has lost both modes of control.

These comments help us understand the intensely traumatic nature of waterboarding. As noted earlier, the survivors in our unpublished study had been subjected to asphyxiation torture 1 to 11 times (average 9). Those who had this experience more than once (up to 11 times maximum in the study sample) were more likely to develop PTSD. This may help better understand the experience of Khalid Sheikh Mohammed – the alleged mastermind behind 9/11 – who was waterboarded 183 times.

In 2007 we published an article in the Archives of General Psychiatry – based on 279 survivors of torture in former Yugoslavia countries – which examined the distinction between physical torture and cruel, inhuman, and degrading treatment (CIDT). The results (reviewed in my previous post) showed that CIDT is not distinguishable from torture in its immediate and long-term psychological impact. As this was the first ever scientific study that investigated this issue, the findings received wide interest from both the human rights community and the world media. A US Justice Department spokesman, when asked by New York Times to comment on the findings, said “…acting with the specific intent of causing prolonged mental harm” would be illegal under United States and international law. This reflected the Bush administration’s view that, since torture causes prolonged mental harm, any ill-treatment not specifically intended to cause such harm is not torture.

It might be of interest to note here that the US Department of Justice memorandum of 2004 had referred to our previous research in support of the criterion of “prolonged mental harm.” Our 1994 study, although showing that torture is associated with PTSD, also demonstrated that 82% of the survivors (mostly resilient political activists) did not have PTSD, despite extremely severe torture. This finding – overlooked by the US Department of Justice lawyers – actually challenged the notion that torture is not torture unless it causes prolonged mental harm (see 2007 article, p. 284, for more discussion of this issue). By clarifying this issue, the 2007 article demonstrated further the scientific untenability of the Bush administration’s definition of torture.

This article was followed by a second one published in the American Journal of Orthopsychiatry in 2009. Based on pooled data from all our studies involving a total of 432 cases (including the 55 cases on which the unpublished manuscript was based), this study took the previous study findings one step further in showing that CIDT is not only indistinguishable from torture in its immediate impact but also causes more mental harm in the long-term. Equally importantly, the study findings pointed to the importance of the contextual processes during captivity / detention in defining torture (see previous post for more details).

So, this is the brief story of our 20-year-long work on “enhanced interrogation techniques.” Sometimes I can’t help wondering what all this work has achieved. Much of this work is in public domain and, in this day and age, it is not difficult to access scientific knowledge on any issue. Yet, we still see arguments of the kind below:

Does “ZD30” [Zero Dark Thirty] glorify torture? No, because no one is tortured in it. The worst procedure shown is waterboarding, and while this is an extremely unpleasant process (it’s not even easy to watch a movie simulation of it), it isn’t torture. Any reasonable definition of torture must exclude procedures that sane people would undergo on a lark. Journalists such as Kaj Larsen and Christopher Hitchens have volunteered to be waterboarded in exchange for nothing more than a cocktail-party anecdote and some copy. (Indeed, Larsen paid $800 for the privilege.) A mixed-martial-arts trainer named Ed Clay volunteered to be waterboarded because he was upset about the general tenor of discussion during a Republican presidential debate and wanted to prove something or other.

In my previous post I noted that the distinction between torture and CIDT in international law reinforces the misconception that the latter causes less harm and might therefore be permissible under certain circumstances. It has been argued that abolishing this distinction would lead to an overly broad definition of torture, making it “harder to maintain the strongly negative aura that torture has.” My 2009 article covers this issue:

…the broader definition of torture implied by these findings cannot be deemed overly inclusive without denying the reality of torture as it is practiced in real life situations. Such views reflect a rather stereotypical image of torture as involving only certain atrocious acts of physical violence. Although such disturbing images might be useful in channeling public reactions against torture, they also foster a skewed image of torture, reinforcing the perception in some people that ‘cruel, inhuman, and degrading’ treatments do not amount to torture. Far from downplaying the problem of torture, our studies highlight the fact that the reality of torture is far more serious than people generally believe.

As noted in an earlier post, I forwarded this point of view, together with the supporting evidence, to the UN Committee Against Torture. I have argued time and again in media interviews that science – not personal opinions or political considerations – should guide us in deciding what constitutes torture. After all, this is very much in keeping with the very essence of international law – as Robin Coupland explains in International Review of the Red Cross:

Whilst there may be ambiguity about the status of humanity in relation to international law, humanity-humankind has been extensively studied by scientific disciplines such as biology, anatomy, physiology, psychology, anthropology and sociology. Health sciences in particular have given us numerous ways to measure humanity-humankind’s well-being or lack of it. Acts of inhumanity or crimes against humanity are all too objective in terms of results, however emotional our reactions to them may be. Such results can, however, be measured in terms of the impact that armed violence or the threat of it have on health. Humanity-humankind and inhumanity can therefore be moved from the ambiguous area where humanity is now situated to the domain of health.… In this way, degradation of personal security and therefore of people’s health can be identified, analysed and commented upon in objective terms before any moral, political or legal judgment is made of the context.

Identifying the impact of degradation of personal security on health in objective terms is indeed what we have done in the last 20 years. Now it is time for a moral, political and legal judgment – one that is not debatable and not open to potential abuse. Until such judgment is made, waterboarding will continue to be not much more than an entertaining topic of conversation at cocktail parties.


Tagged as: , , , , ,

2 Responses »


  1. Torture does not work, says trauma expert - Bioethics Research Library
  2. An unextraordinary claim: Torture doesn’t work | Canadian Atheist

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Enter your email address to follow this blog and receive notifications of new posts by email.

%d bloggers like this: