The CIA once trained cats to be Cold War spies
Here’s the scene: A man in a trench coat and fedora sits on a park bench, frequently looking up from his newspaper to sneak glances at passersby. A stray cat passes by. She rubs against the man’s legs. He pets it absentmindedly and continues to look anxiously at his surroundings.
Finally, another man arrives, carrying a briefcase. He sits next to the man with the newspaper. They exchange a few words in Russian, then the second man leaves, leaving the briefcase behind. The first man sits for a moment longer, then picks up the briefcase and walks off in the opposite direction.
We’ll never know what nefarious plans these two were up to…or will we? Could this adorable feline really have collected relevant information?
[dramatic music]
Well no, that’s not going to happen. Despite the CIA’s best efforts during the Cold War, the use of cats as spies became, unsurprisingly, a disaster.
Keeping cats for national security
The 1960s were a crazy time for the CIA. When they weren’t shooting themselves with acid or trying to use explosive cigars to kill Fidel Castro, Agency personnel were exploring other new approaches to espionage, like Project Acoustic Kitty.
The mission was to use the domestic cat, the most famous of cooperative, obedient and not even slightly contrary animals, to try to collect information from the Soviet embassy.
Using animals to spy on the embassy may or may not have been a viable idea, but the CIA made things terribly difficult for itself by choosing an animal for the job.
Unsurprisingly, the Acoustic Kitty program was a failure, largely because – in what will shock cat owners to the core – the cat chosen for the mission refused to do what it was supposed to do.
“We never found an animal we couldn’t train.”
It’s unclear exactly what went wrong, largely due to conflicting accounts from the two primary sources on the project. The first of these is former CIA agent Victor Marchetti, who discussed the program in an interview with British documentarian Adam Curtis in the latter’s film. You’ve used me as a fish long enough. The second is animal trainer Bob Bailey, who spoke to Smithsonian magazine about the program in 2013.
Marchetti and Bailey agree that the cat was essentially transformed into a walking radio, with a transmitter implanted in its abdomen.
Marchetti claims that during his first mission, the poor cat was hit by a car before approaching his targets.
Bailey counters this and maintains that although it seems absurd – and frankly cruel – the project was not a complete failure. “We never found an animal we couldn’t train,” he said. “We found that we could condition the cat to listen to voices…we found that the cat listened more and more to people’s voices and less and less to other things.” Acoustic Kitty, he asserted, was “a serious project”.
The only other source we have on the project is a CIA document titled “Views on Trained Cats,” which is accessible – in heavily redacted form – through the National Security Archives at George Washington University.
The document states that “it is indeed possible…”, but the rest of the sentence is redacted. The next sentence proclaims that whatever is actually possible is also “a remarkable scientific achievement” and that “the work done on this problem over the years is a credit to the staff who guided it.” (Staff are not named.)
Ultimately, however, the paper concludes that even if “cats can indeed be trained to move short distances,” the whole thing is a dead end: “The program would not lend itself, from a practical standpoint, to our highly specialized needs.” »
Why are cats so difficult to train?
Anyone reading this project might sympathize with cats who don’t care at all for our highly specialized needs. But still: why are Are cats so reluctant to do everything we want? After all, dogs are desperate to please. Why are cats so…difficult?
Stephen Quandt, a professional cat behaviorist based in New York, tells Popular science that there is an essential difference between man’s best friend and the humble domestic cat: “Dogs were bred to want to please us. Retrieving a ball feels good to them, but it’s also good that we ask them to retrieve it and they are able to do it.”
Cats, on the other hand, don’t particularly care whether what they do makes us happy, as long as it makes us happy. them happy: “Cats want to do exactly what they like, regardless of whether we want them to do it or not. »
It may well have something to do with the way cats and dogs were domesticated. Although there are competing theories about exactly how animal domestication took place, it is widely accepted that cats self-domesticated. They came into contact with humans when they started hunting vermin that lived in our attics, and we let them stay because killing that vermin was also useful to us.
Dogs, on the other hand, have been deliberately bred for a wide variety of purposes, from herding sheep to retrieving birds to helping the blind. Training dogs to perform these tasks requires extensive interaction. Cats, on the other hand, had only one job: killing vermin – and it was something they did with enthusiasm anyway.
Related Stories “In That Time When”
How to Train a Cat to Become a Cold War Spy
With all of this in mind, it’s no surprise that the cat chosen for Acoustic Kitty showed very little interest in spying on anyone, Soviet diplomats or otherwise. Fortunately, the days of cloak-and-dagger feline agents are probably over: these days it would be just as easy to use a small drone or something. But when it came to training a cat to spy on a Soviet spy, how would he go about it?
“I’d probably try to find a really friendly cat,” he laughs. He explains that ideally the cat would be familiar with the people in question. Since that would be impossible in this scenario, he would fall back on the universal motivator: food. “If the spies were outside, maybe they would have a picnic sometimes. So I was trying to make sure the cat was hungry.”
What is the best way to train a cat?
More generally, though, he says there is a common method for training cats who perform in movies, circuses, etc., and that involves a device called a clicker. “If you give a cat a treat when it’s doing what you want it to do, its walnut-sized brain makes the connection and it says, ‘Oh, a treat!’ »
If you accompany the treat with a sound, the cat begins to associate the sound with the treat. This classic Pavlovian scenario is called “click and treat.” As Quandt explains, “The click becomes an indicator of a reward, and eventually it itself becomes a temporary reward. The cat knows the treat is coming.”
Once the cat associates the clicker with a reward, it becomes reasonably simple to establish an association between that reward and a desired behavior. Scratch the scratching post instead of the sofa? Click. Refrain from attacking your friend while he’s peeing? Click. Surreptitiously retrieve nuclear codes from the man in the trench coat reading a newspaper upside down? Click.
That’s how we end up with cats like Internet favorite Owlkitty, whose appearances in big-budget feature film parodies are the stuff of YouTube legend. Was it also used in the formation of Acoustic Kitty? We will never know – and I hope we never have any reason to find out again.
In This time whenPopular Science tells the strangest, surprising, and little-known stories that have shaped science, engineering, and innovation.
The best new PopSci 2025 releases
