How US Intelligence Agencies Hid Their Most Shameful Experiments

Matthew Connelly on Human Cost of Secret Science

Although proponents of secret science like to focus on examples in which it has benefited society, insiders from the very beginning of the Cold War worried that the best minds would not be drawn to work that they could not even talk about. Secrecy protected those involved from embarrassment or criminal prosecution, but it also made it much harder to vet experimental protocols, validate the results, or replicate them in follow-up research.

One research manager at a Department of Energy weapons lab would later admit, “Far more progress is actually evidenced in the unclassified fields of research than the classified ones.” The physicist Robert McCrory, whose own lab received millions in funding in partnership with Lawrence Livermore, Sandia, and Los Alamos National Laboratories, was even more blunt: “Some of the work is so poor that if it were declassified, it would be laughed off the face of the Earth.”

We can only guess what, specifically, McCrory had in mind when he said this. There are all too many possibilities. Collectively, they lend credence to the oft-stated concern that secret programs became a refuge for second- and third-rate minds. The wizards of Langley, for instance, considered it a “remarkable scientific achievement” when they managed to prove that cats could be “trained to move short distances.” According to a CIA veteran, Victor Marchetti, this achievement was part of a program to determine whether cats could be turned into surveillance devices:

A lot of money was spent. They slit the cat open, put batteries in him, wired him up. The tail was used as an antenna. They made a monstrosity. They tested him and tested him. They found he would walk off the job when he got hungry, so they put another wire in to override that. Finally they’re ready. They took it out to a park and pointed it at a park bench and said, “Listen to those two guys. Don’t listen to anything else—not the birds, no dog or cat—just those two guys!” They put him out of the van, and a taxi comes and runs him over. There they were, sitting in the van with all those dials, and the cat was dead!

The CIA nevertheless commended the “energy and imagination” of the team, and considered them potential “models for scientific pioneers.”

Secrecy protected those involved from embarrassment or criminal prosecution, but it also made it much harder to vet experimental protocols.

It could be argued that a sprawling research program purposely designed to push the envelope will, over several decades, inevitably produce some strange and low-quality research. But in some cases it is possible to make a side-by-side comparison of US government research with research commissioned by another country that had fewer resources but the same goal.

For instance, during World War II, American and British forces both used dogs to detect mines. It was delicate, dangerous work, and the dogs sometimes proved unreliable. Both governments therefore mounted research projects in the early 1950s to evaluate and improve dogs’ ability to locate mines.

The British just wanted “the facts” and sought out a “trained scientist.” They selected Solly Zuckerman, an anatomist expert in animal behavior. He designed the experiment to eliminate the possibility that human handlers were unconsciously influencing the dogs’ performance.

This required systematically isolating the specific biochemical and physiological factors that might explain success or failure, since either could prove important when mines were odorless. Zuckerman had a strong personal motivation—he had seen the devastating impact of blast injuries when he conducted wartime physiological research with the survivors. His larger agenda was to develop more rigorous experimental methods in animal research. Zuckerman found no solid evidence that dogs could be relied on to detect buried mines.

The U.S. Army, on the other hand, hired a “parapsychologist” named J. B. Rhine. It is not clear why—all the army records were later destroyed. Though his training was in botany, Rhine had become famous for his experiments—never replicated— in extrasensory perception (ESP) and psychokinesis. For Rhine, the study was an opportunity to prove that ESP really existed; he had already gathered a collection of amazing stories of animal ESP. Rhine once again convinced himself that dogs possessed special powers.

Army officials found in follow-up work that the results were random, and another study proved to be a complete failure, marked by a “rather conspicuous refusal of the dogs to alert.” But Rhine used the army money to seed new research, and found new customers. The Office of Naval Research funded a decade of work on ESP in homing pigeons. In other studies, one of Rhine’s colleagues tried to influence a cat telepathically to select one dish of food over another. But here, too, even Rhine admitted that the results were “not spectacular.” Alas, the cats proved “elusive.”

The US government would spend several decades on the larger program of mind-control research. And Rhine was a paragon of scientific rigor compared with some of the other researchers on the government payroll, who espoused theories of extraterrestrial and ghostly visitations to explain ESP, and were hired by the US Army to consult on psychedelic mushrooms.

The CIA’s Project MKUltra involved a whole series of experiments on unwitting subjects, using a range of different drugs in order to manipulate them into saying and doing things against their will. It was given carte blanche to operate without the normal Agency accounting controls or need for written contracts. Once again, researchers rapidly escalated their trials with little understanding of the effects. In the first round of an experiment conducted at the New York State Psychiatric Institute, one of the patients, Harold Blauer, was given 0.4 mg of methylenedioxyphenyl-isopropylamine, a drug similar to ecstasy. The next dose was sixteen times stronger, and Blauer was dead in thirty minutes.

When the family took legal action, government lawyers threatened witnesses with prosecution under the Espionage Act. Decades later, CIA Director Stansfield Turner admitted that “some unwitting testing took place,” but testified to Congress that the subjects were “criminal sexual psychopaths confined at a State hospital.” In fact, Blauer was a tennis pro who voluntarily sought treatment for depression after a divorce.

The CIA also experimented on its own personnel. In one case, the head of MKUltra, a chemist named Sidney Gottlieb, dosed the attendees of a joint Agency-army retreat with LSD. One of the unwitting subjects, an army biochemist named Frank Olson, was traumatized by the experience. Until then, Olson had been an outgoing and devoted family man. Afterward, he sank into depression, overcome with feelings of shame, and would not return home. He told his army supervisor that he wanted to quit or be fired.

Gottlieb was likely alarmed upon being warned about this situation. MKUltra had been approved by the CIA director himself, Allen Dulles, who called the program “ultra-sensitive.” But Gottlieb reportedly had not obtained prior authorization before drugging Olson and the others. As an internal Agency report later noted, participants in this work well understood that their methods were “professionally unethical” and legally dubious, and would provoke “serious adverse reaction” from the public if ever revealed. Gottlieb therefore had a powerful motivation to make certain Olson told no one about what the CIA had done to him.

The government’s venture into the paranormal proved useless for any legitimate intelligence purposes.

Gottlieb and his deputy decided to take Olson to see a New York doctor. The man had no psychiatric training but did have a top-secret CIA security clearance and experience with LSD. The doctor plied Olson with bourbon and sedatives, and took him to see a performance by a magician, whom Gottlieb was interested in hiring to help dose more unwitting targets. Olson’s public behavior became increasingly erratic, and he said that the Agency was “out to get him.” The doctor said he would take Olson to a sanitarium to be treated by CIA psychiatrists. But that night, Olson “fell” from the tenth floor of the Statler Hotel in Manhattan.

Gottlieb’s deputy was sharing the room with him, and claimed to have been asleep when it happened. His story was that Olson had just crashed right through the window, without opening it first, or even raising the shade. We may never know what happened in that hotel room. But at the time, the CIA was training its assassins to first ply their victims with drugs or alcohol, and advised, “The most efficient accident, in simple assassination, is a fall of 75 feet or more onto a hard surface.”

The CIA was clearly prepared to kill innocent American citizens. Six weeks after the Agency covered up what happened to Olson, a CIA mind-control team went on its first foreign assignment. The mission was to slip “artichoke”—likely LSD—into the drink of another unwitting subject and induce him to attempt an assassination of a prominent politician or American official. The artichoke team was primed for action, proudly noting that they “were ready when called upon for support, even though the operation did not materialize.”

Most of the MKUltra records were later destroyed, so we cannot know what other missions might have aimed to achieve. But CIA mind-control research went on for more than a decade, involved some eighty different institutions, and would eventually cost approximately ten million dollars—about a hundred million in today’s dollars.

A whole new mind-control program started up in 1972, this time led by the Defense Intelligence Agency (DIA). The goal of this new program was “to determine whether anomalous mental phenomena (i.e. extrasensory perception and psychokinesis) existed and the degree to which such phenomena might be applicable to problems of national interest.” Like MKUltra, it went on for many years, and cost many millions of dollars. How many exactly is still difficult to determine. But a single California contractor, SRI International, would eventually receive $11.3 million (or about thirty-six million in today’s dollars).

Even escape artists and magicians saw that the government was being rooked, and patiently explained to officials how timeworn tools of their trade could easily trick someone into believing in ESP. So, too, did the scientists at DARPA, who concluded that the Israeli illusionist Uri Geller, the DIA’s star pupil, was a “charlatan.” They thought it was “ridiculous” that Geller had fooled the US government into using taxpayer dollars to see whether he could bend spoons with his brain. They pointed to a host of problems with the ESP and psychokinesis experiments, above all the fact that the people paid to conduct them had a financial incentive to produce positive results.

Under the cloak of secrecy, their hubris and magical thinking ran wild.

In 1985, the army commissioned a blue-ribbon panel from the National Research Council to evaluate the program. The panel concluded there was “no scientific warrant for the existence of parapsychological phenomena” such as “remote-viewing”—sensing the location or appearance of things through sheer mental effort—or psychokinesis.

Nevertheless, over the following decade the army conducted between fifty and a hundred more such experiments. In 1995, another review of the remote-viewing program was commissioned, this time by the American Institutes for Research (AIR). Once again, the reviewers found that, because of flaws in the research designs, there was no clear evidence demonstrating the existence of the paranormal.

But the AIR report found something even more damning. After some twenty-five years of experiments, the reviewers concluded, “In no case had the information provided ever been used to guide intelligence operations.” Even if some people really do have ESP that cannot be explained by science, the point of the program was not to use government resources to explore the Twilight Zone. It was to support actual missions that would safeguard national security. Yet, despite all the time and money spent—not to mention the human costs—the government’s venture into the paranormal proved useless for any legitimate intelligence purposes.

Why, then, did the intelligence community and the Pentagon go to extremes in pursuing such embarrassing “research”? For the same reason why they felt they had license to control the weather and alter the upper atmosphere: because, under the cloak of secrecy, their hubris and magical thinking ran wild. Moreover, controlling people’s minds was a prize that was just too tempting to resist. And although the government may have given up on telekinetic spoons, it did not give up on that larger goal.

During the first decade of the “Global War on Terror,” the CIA pursued mind control through more direct methods—i.e., “enhanced interrogation.” The program employed psychological abuse, stress positions, and waterboarding not just to make people talk, but also to discover scientifically rigorous and reproducible methods for compelling subjects to submit to the will of interrogators and lose all sense of personal agency. Just like MKUltra, the enhanced-interrogation program was conducted as a series of “experiments.” Here is how the CIA’s Office of Medical Services, in a top- secret 2004 document, described the protocol for recording the application of “treatments” to “subjects”:

In order to best inform future medical judgments and recommendations, it is important that every application of the waterboard be thoroughly documented: how long each application (and the entire procedure) lasted, how much water was used in the process (realizing that much splashes off), how exactly the water was applied, if a seal was achieved, if the naso- or oropharynx was filled, what sort of volume was expelled, how long was the break between applications, and how the subject looked between each treatment.

The CIA hired a retired air-force psychologist named James Mitchell to enact these methods. Mitchell, too, saw himself as a scientist. As an informed source recounted to the journalist Jane Mayer, after Mitchell took over a case, he told the FBI agents that an interrogation “was like an experiment, when you apply electric shocks to a caged dog, after a while, he’s so diminished, he can’t resist.” When the agents argued that the subject of this experiment was a human being and not a dog, Mitchell retorted, “Science is science.”

Waterboarding, “diapering,” and stress positions ultimately proved no more effective than ESP or psychokinesis, as the CIA’s own internal reviews concluded. Many of the victims had already provided valuable intelligence to their interrogators before they were tortured, and once the waterboarding began, many offered false information in order to make it stop. But the waterboarding did not stop until Mitchell’s company was paid eighty-one million dollars.


Excerpted from The Declassification Engine: What History Reveals about America’s Top Secrets by Matthew Connelly. Copyright © 2023. Available from Pantheon, a division of Penguin Random House, LLC. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.

Matthew Connelly
Matthew Connelly
Matthew Connelly is a professor of international and global history at Columbia University and the principal investigator at History Lab, an NSF-funded project to apply data science to the problem of preserving the public record and accelerating its release. He received his B.A. from Columbia and his Ph.D. from Yale. His previous publications include A Diplomatic Revolution: Algeria’s Fight for Independence and the Origins of the Post-Cold War Era, and Fatal Misconception: The Struggle to Control World Population.

More Story
Chetna Maroo: “Every Time I Lost my Way, I Would Go Back to That First Page.” Chetna Maroo is the guest. Her debut novel, Western Lane, is out now from Farrar, Straus, & Giroux. Subscribe and download...