In May 1943, a New York gangster known as Little Augie was handed a cigarette. He didn't know it was laced with marijuana. Little Augie was a hard case — he prided himself on never having been an informer, and he'd reportedly helped kill people who were informers. He'd been uncommunicative about his crimes. After two of those cigarettes, he started talking. He named names. He revealed that he'd been bribing a drug-enforcement official for years, that an associate was bribing a restaurant inspector, that two law enforcement officials were shaking down a liquor businessman.1
This was a US government experiment — run by a formal committee of distinguished academics and intelligence officials convened by the Office of Strategic Services, the wartime forerunner of the CIA. And the lesson the committee drew from Little Augie's loosened tongue is the one that would go on to poison the next two decades: the drug seemed to work if administered surreptitiously. That single conclusion — dose them without their knowledge — is the seed of Operation Midnight Climax, Frank Olson's death, and the whole catastrophic CIA LSD program.2
The striking thing about the OSS truth-drug committee is how legitimate it looked. This wasn't a rogue operation. It was a panel convened through the National Research Council, chaired by Dr. Winfred Overholser (professor of psychiatry at George Washington, director of St. Elizabeth's Hospital), and staffed with figures like Dr. Lawrence Kubie (a prominent psychoanalyst) and a Johns Hopkins psychiatry professor.3 Serious institutional science, harnessed to the problem of extracting information from enemy prisoners.
And then there are the two names that tell you where this was heading. One is Harry Anslinger — the first commissioner of the Federal Bureau of Narcotics, America's most relentless anti-marijuana crusader, the man who called marijuana the "assassin of youth" and warned it had "a record of murder and terror running through the centuries." He sat on a committee that was tentatively endorsing marijuana as an interrogation drug. The irony is almost too neat: the nation's chief marijuana prohibitionist, lending his name to a program using marijuana to loosen tongues.4
The other name is George White — an intelligence agent, a late addition, the operational muscle among the academics. White shared his field skills with the panel's professors, and he would go on to run some of the CIA's most notorious LSD experiments, including the prostitute-run drugging operation on Telegraph Hill. The committee is where the respectable science and the rogue operative first sat at the same table — and the operative's instincts won.
Walk through what the committee actually found, because the pattern is the same one Plötner found at Dachau and the Yale investigators found in the lab, and the committee's response to it is the tragedy.
They tried mescaline on three volunteer officers: it didn't induce relaxation or make them divulge anything. They tried marijuana by mouth in liquid drops: physical discomfort, no disclosure. They sprayed marijuana vapor into a room, then moved to cigarettes.5 With the cigarettes, they thought they saw promise — the drug seemed to bring on a "state of irresponsibility" in which a subject became loquacious and freely shared information. And then came Little Augie, the field test on a real, resistant criminal, who talked after two cigarettes.
But look at the caveats the committee itself recorded. There were wide individual differences in susceptibility, so dosage was critical. The effect was unreliable person to person. And the drug's "success" depended entirely on the subject not knowing he was drugged — hence the conclusion that it worked if administered surreptitiously.6
Here is the poison in that conclusion. The committee had actually demonstrated that the drug didn't reliably compel truth (mescaline failed, oral marijuana failed, susceptibility varied wildly). What it had shown was that a disinhibitor, slipped to an unsuspecting person, sometimes made them chatty. But rather than reading that as "the truth-drug idea doesn't work," the committee reframed it as "the truth drug works, if you administer it in secret." That reframe converted a negative result into a mandate for covert dosing of unsuspecting people — the exact practice that would maim and kill Americans over the next twenty years.
The word "surreptitiously" did enormous damage. It took a failed hypothesis (drugs compel truth) and rescued it by adding a condition (as long as the subject doesn't know), and that condition became a license. If the drug only "works" when the subject is unaware, then the program requires dosing people without consent — and so an experiment that should have ended the truth-drug dream instead chartered the era of secret drugging.
What makes this committee worth its own page is that it's the institutional origin of the American coercive-drug enterprise — the moment the fantasy acquired official standing, respectable sponsors, and a foundational (wrong) conclusion. Everything downstream flows from here: the surreptitious-administration doctrine, the involvement of legitimate academics, the presence of George White, the willingness to test on non-consenting subjects (Little Augie never agreed to be drugged).
And notice the structural pattern, because Dimsdale draws it deliberately: a serious scientific committee, working on a real national-security problem, produces a conclusion shaped less by its own evidence than by what the operational people wanted to hear. The academics ran experiments that mostly failed; the intelligence agent's instinct — dose them secretly — is what survived into doctrine. This is the template for MKUltra: legitimate science captured and steered by an agency agenda, its negative results laundered into operational green lights.
You chair a wartime committee tasked with finding a drug to make enemy prisoners talk. You have real scientists and one intelligence operative. Watch how you arrive at exactly the wrong conclusion.
You test methodically. Mescaline on volunteer officers — nothing. Oral marijuana — discomfort, no disclosure. You keep going, because the mission is urgent and the sponsors want a yes. You try vapor, then cigarettes, and with the cigarettes you think you see something: a loosening, a loquacity. You run a field test on a real gangster who doesn't know he's dosed, and he talks.
Now you have to interpret a messy result: the drug failed in most forms, susceptibility varied wildly, and it only seemed to work when the subject was unaware. The honest reading is this doesn't reliably work. But the mission pressure and the operative's instincts push you toward the rescue clause: it works if administered surreptitiously. You write that down.
What tells you you've just done real damage is the shape of the conclusion — you've taken a hypothesis your own evidence undermined and saved it by attaching a condition (secret dosing) that then becomes a mandate. You haven't proven the drug compels truth. You've chartered a program to drug unsuspecting people, indefinitely, in pursuit of a capability your experiments already suggested doesn't exist. The word "surreptitiously" has turned your failure into someone's license.
The evidence is unusually concrete: the declassified OSS documents (Dimsdale reproduces a June 1943 memo cover page), the committee's membership roster, the recorded results of each drug trial, and the Little Augie field-test report. This is the paper trail of the program's origin, in the government's own files.
The tension the book leaves open is the one that recurs through every coercion-research chapter: were the committee's scientists naïve, captured, or complicit? They ran genuine experiments and recorded genuine caveats — that's real science. But they let the "administer surreptitiously" conclusion stand, and that conclusion required abandoning informed consent and chartered decades of harm. The unresolved question is where the responsibility sits: with the operative whose instinct won, with the academics who lent legitimacy and didn't stop it, or with the institutional structure that reliably converts urgent missions into laundered green lights for whatever the agency already wanted to do. Dimsdale's answer, developed across the MKUltra chapters, is "all of the above, and the structure most of all."
This committee is the Allied twin of The Nazi Doctor Who Spiked the Coffee — the two wartime truth-drug programs, on opposite sides, reaching opposite conclusions from the same evidence. Plötner honestly wrote "it is impossible to impose one's own will"; the OSS committee, facing the same failure, rescued the fantasy with "if administered surreptitiously." The comparison is damning: the Nazi accepted the negative result the Americans refused to accept, and it was the Americans' refusal — not the Nazi's honesty — that chartered twenty more years of drugging.
It is the direct ancestor of MKUltra Institutional Architecture and specifically of George White and Operation Midnight Climax. White is at this table in 1943; the surreptitious-administration doctrine written here is the doctrine he executes in San Francisco a decade later. The committee is where MKUltra's founding error and one of its key operatives were both already present.
To behavioral-mechanics — Surreptitious Drugging as Control Vector. This committee is the documentary birthplace of surreptitious drugging as official American doctrine. The insight the pairing produces: the "administer surreptitiously" conclusion wasn't a finding about what works — it was a rescue operation for a failing hypothesis, and it worked by relocating the drug's unreliability out of view. If the drug only "works" on the unaware, then you can never cleanly test it (an aware subject is a failed test by definition), which means the hypothesis becomes unfalsifiable and the program becomes permanent. Surreptitious administration didn't make the drugs effective; it made their ineffectiveness un-provable, and an un-provable failure is a research program that never has to end. That's the mechanism by which a 1943 committee's messy negative result became a self-perpetuating twenty-year catastrophe.
To psychology — The Ordinary Person Thesis. The committee's own recorded caveat — "wide individual differences in susceptibility to the drug, and thus dosage was critical" — is an early, inconvenient data point against the dream of a universal chemical lever. People vary; there's no dose that works on everyone; the same drug that loosens Little Augie leaves the next man silent or sick. The connection reveals the tension the truth-drug program never resolved: coercive persuasion's most reliable methods (sleep deprivation, isolation, dread) work on nearly anyone because they exploit universal human physiology, while the drugs work unpredictably because they hit idiosyncratic individual chemistry. The committee wanted the reliability of the DDD approach with the speed of a pill, and the "wide individual differences" caveat is the quiet admission that they couldn't have both — that the search for a universal truth drug was, from the start, chasing a consistency that pharmacology couldn't provide.
Sharpest implication: A committee of respectable scientists demonstrated that truth drugs mostly failed — and then rescued the fantasy with one word, "surreptitiously," converting a negative result into a mandate for dosing unsuspecting people that would run for twenty years and kill Americans. The rescue clause was diabolical because it made the failure unfalsifiable: if the drug only "works" on the unaware, no honest test is possible, and an un-disprovable failure is a program that never ends. The Nazi at Dachau accepted the negative result; the American committee refused it — and the refusal, not the science, is what built MKUltra.
Generative questions: