"How about when you're, um, coughing?" researcher Penny Patterson asks from off screen.
Koko the gorilla raises a hand to her mouth, waits a beat, then wheezes into it, sounding every bit like an aging smoker.
"That was good!" Patterson cheers, while Koko holds a massive hand out to her, as though accepting the praise.
The video is one of dozens posted on the Gorilla Foundation Web site. The California non-profit has footage of its celebrity ape playing a recorder, blowing her nose into a handkerchief, babbling into a toy phone held up to her head in the crook of her right elbow.
The clips fit neatly into the foundation's savvy campaign to humanize animals like Koko, who according to the foundation's Web site has a 1,000-word sign language vocabulary and loves the movie "Pretty Woman."
But according to new research, Koko's ability to cough on command is not just a neat party trick or a manufactured heartwarming moment. It might help rewrite our narrative of what great apes — humans' closest living relatives — are capable of. And in doing so, it may teach us more about our own history of communication.
For a recent study in the journal Animal Cognition, language researchers Marcus Perlman and Nathaniel Clark mined Gorilla Foundation recordings to identify nine distinct "vocal and breathing behaviors," that Koko has learned from more than 40 years living with humans, including coughing, blowing into the recorder and "speaking" into the phone.
Unlike most sounds gorillas are known to make, these are not reflexive responses — like a human's yelp after putting a hand on a hot stove. And they're not part of a gorilla's normal repertoire of noises, the sounds they instinctively know how to create. Rather, they're skills Koko has acquired through training, exerting control over her voice and breathing to make the sounds at will.
Perlman is clear that the vocalizations are not the same thing as speech. But they're still significant, he said, because gorillas aren't supposed to be able to make these sounds at all.
"There's a longstanding idea that non-human primates they have an extremely limited ability to learn new vocalizations and to control those vocalizations voluntarily," Perlman, a cognitive psychologist at the University of Wisconsin and the paper's lead author, told The Washington Post. "But the Koko case study shows that's not true. She was able to acquire new vocal behaviors and also breathing, and that suggests there's a basic capacity in place with great apes, that maybe this capacity over millions of years evolved into what we now know as speech."
The belief that great apes — gorillas, bonobos, chimpanzees and orangutans — are incapable of anything even resembling speech dates back to the mid-20th century. In 1951, after twenty years of raising a chimpanzee as if it was their child, psychologists Keith and Cathy Hayes concluded that apes could never be taught to talk.
"We realized an astonishing fact," Cathy Hayes wrote in the introduction to her account of the experience, "The ape in our house." "Before our coaching Viki [the chimp] had been completely unable to make any sound at all on purpose. She made chimp noises, yes, but these were beyond her control. They were merely reflex expressions of her feelings. . . . She had lacked the motor skill of vocalization."
Later, scientists would learn that the mouths, wind pipes and vocal cords of apes are fundamentally different from those of humans. Our tongues are more flexible, our lips more sophisticated, our lungs specially designed to control the strength of our breath. Beyond that, our brains are just better equipped to control those parts of our bodies. Our Broca's area — the part of the brain linked to language processing and speech — is much larger, and our neurons more significantly connected to our vocal tract.
So researchers switched to sign language for their ape experiments, resigned to the idea that apes' bodies and brains just weren't designed for speech.
But those discoveries left researchers with an unexplained gap in the evolutionary history of language. Other kinds of more distantly related primates, like monkeys, are known to produce human-like "precursors" to speech: chitters, chortles, harmonic tones. And most humans in most environmental circumstances have developed spoken language, even without being explicitly taught. Yet the great apes, related to humans even more closely than monkeys, were supposedly speechless.
The consensus was that vocal communication as we know it today developed solely along the human line, after we had split off from our primate relatives. Whatever precursor to language may have existed was probably restricted to gestures, where — as we know from Koko and other animals' extensive sign language vocabulary — apes have much more developed capabilities.
But according to Perlman, Koko's ability to learn new vocalizations "changes that narrative." She may not have the same vocal control as humans, but she has enough to make her voice and her breath do what she wants.
"If some basic capacity was there at the time of our last common ancestor . . . it suggests that language evolved maybe through some kind of combination of gestures and vocalization," he said. "There's still a gap between apes and human speech, but the gap is smaller than we previously thought."
Koko is not the only great ape to show the ability to make these human-like sounds, though she's the first gorilla to do so. Earlier this year, a report in PLOS One cataloged similar vocalizations, including clicks and a Sam Elliot-like warble, from a Swiss orangutan named Tilda. And just two weeks ago a study in the journal Peer J concluded that bonobo "peeping" was the first known instance of "structural flexibility" in animals — the ability to interpret the meaning of acoustically identical vocalizations differently based on context.
"The more we look, the more continuity we find among animals and humans," lead author Zanna Clay of the University of Birmingham's School of Psychology said in a statement to The Washington Post.
Rob Shumaker, a co-author of the orangutan study and vice president of the Indianapolis Zoo, praised Perlman's paper as a "very fine contribution" to the growing body of research on apes and speech.
"It always stuns me how the assumption [that apes have no ability to control their vocalizations] just caught on," he told The Post. "Yet the differences between humans and great apes," he paused for a moment, searching for the right phrasing, "these differences are really in degree, not in kind."
In Shumaker's view, the belief that apes weren't part of the evolution of spoken language "was always about preserving human uniqueness," he said. "But humans are not as unique as we think."
There's still some skepticism in the scientific community about that last point. Studies have challenged the idea that apes understand the meaning of their sign language communication the way humans do, or suggested that famous primates like Koko, who have lived their whole lives in the company of humans, aren't representative of their species.
But Perlman is a language researcher, not a primate specialist. He's not trying to convince anyone of the potential for inter-species connection between apes and humans. What interests him is not what Koko may or may not be trying to say, but what the fact of her babbling tells us about language's ancient, complex origins.
That our closest animal relatives have this capacity, "fits with the view that human language use is a profoundly multimodal activity that combines speech, gesture, facial expression and other bodily postures within a tightly integrated and synchronized system of expressive movements," he writes in his study.
In some ways, this view of the evolution of speech is more interesting than the idea that it started solely with humans. The way Perlman writes about it, vocal communication is almost a dance, one we've been slowly choreographing and complicating for millions upon millions of years.
Perhaps it's about time a coughing gorilla was invited to the party.
