In 1917, when the field of psychology was young and struggling to gain acceptance in science, the American Psychological Association (APA) needed a friend. Like many at the time, it decided to assist the war effort by working with the US military. The collaboration was largely benign: efforts to assess which recruits were fit to be soldiers led to the first formal study of variation in human intelligence. Later, psychologists studied the effects of war on soldiers returning home, fuelling the case for making the First World War “the war to end all wars”.

That was not to be, but psychology, and the APA in particular, continued its close bond with military and intelligence agencies. The relationship is not inherently problematic: indeed, the US Department of Defense (DOD) spends tens of millions of dollars each year on research into post-traumatic stress disorder and other psychological and psychiatric complications of war. The DOD, which employs around 700 psychologists, was a key ally in psychologists obtaining the authority to write prescriptions in the 1980s. And the APA has at times taken a stand against DOD policies: as early as 1991, the organization protested against the Pentagon’s policy of stopping openly gay people from serving in the military by banning DOD advertisements in APA publications.

Nevertheless, the tone of the alliance between US agencies and psychologists has darkened over the past century. Most famous is the Central Intelligence Agency’s (CIA) mind-control programme MKUltra during the cold-war era, in which psychologists helped the CIA to develop and test interrogation techniques involving the use of hallucinogenic drugs and hypnosis.

Given this history, it should be no surprise that the APA has continued to cultivate a close relationship with the agencies. Last week, a long-awaited external report confirmed suspicions of the APA’s involvement in the torture of detainees following the terrorist attacks of 11 September 2001 and the ensuing ‘war on terror’ (see go.nature.com/4vpdob). Starting in 2005, the report found, APA officials worked with the DOD to keep the organization’s ethics guidelines loose enough to justify the participation of psychologists in the DOD’s ‘enhanced interrogation’ programme. As a result, the DOD and CIA could easily brush aside the ethical concerns of their own psychologists: the APA had given the programme its imprimatur.

The story is rife with conflicts of interest: according to the report, six of the nine voting psychologists on the APA task force that wrote the guidelines had consulting relationships with the DOD or CIA, and one former APA president owned a financial stake in the consulting company that oversaw the CIA interrogation programme. As criticism surfaced, the APA defended itself by formally condemning torture while doing nothing to stop its members from participating. Meanwhile, Guantanamo Bay’s chief military psychologist told an APA meeting: “If we removed psychologists from these facilities, people are going to die.” It is an assertion that does more to reveal the disgraceful state of the programme than to offer a moral defence.

There is no evidence that the United States gained any useful information.

Not only did APA psychologists deem the torture programme ethical, but they also gave it a patina of legitimacy by trying to cast it as research. The “studies” — which violate every consent rule for human subjects, including the CIA’s own — involved questions about the acceptable limits of human suffering and how well various techniques could yield useful information from a prisoner. There is no evidence that the United States gained any useful information in this way.

The scientific basis for the interrogation programme was questionable from the start. The theory of ‘learned helplessness’ was developed to test psychiatric drugs by measuring how long mice will try to swim in a bucket of water — depressed animals will give up sooner and allow themselves to be rescued. The psychologists who developed the CIA’s interrogation techniques reversed this idea, theorizing that simulating the experience of drowning, or waterboarding, could induce despair in human detainees until they gave up their story.

The APA has apologized for its failings and has indicated that it will revise its policies to prohibit psychologists from participating in military interrogations. It has also parted company with its ethics director, who the report named as leading the collusion with the military. More heads are likely to roll, and some psychologists could even face prosecution.

The American Psychiatric Association and the American Medical Association forbade their members in 2006 from participating in the interrogation programmes. This is in keeping with the Geneva Convention, an international agreement signed in 1929 and revised nearly 70 years ago to do away with torture and abusive experimentation on prisoners of war. The APA deserves all the criticism it receives and more, for its willingness to forswear global consensus in the interest of making a deal with the devil.