January 19, 2017 at 7:00 AM
"Oh Long Johnson," a cat once said, back in the primordial history of Internet memes. "Oh Don Piano. Why I eyes ya."
Or so said the captions — appended to the gibberish of a perturbed house cat on "America's Funniest Home Videos" in 1999 and rediscovered in the YouTube era, when millions of people heard something vaguely human echo in a distant species.
It was weird. And hilarious. And just maybe, profound.
As the "Oh Long Johnson" craze was fading a few years ago, a wave of scientific discoveries about apes and monkeys began upending old assumptions about the origins of language.
Only humans could willfully control their vocal tracts, went the established wisdom. Until Koko the gorilla coughed on command.
Surely, then, our vowels were ours alone. But this month, researchers picked up British ohs in the babble of baboons.
Study after study is dismantling a hypothesis that has stood for decades: that the seeds of language did not exist before modern humans, who got all the way to Shakespeare from scratch.
And if so much of what we thought we knew about the uniqueness of human speech was wrong, some think it's time to take a second look at talking pet tricks.
"It's humbling to understand that humans, in the end, are just another species of primate," said Marcus Perlman, who led the Koko study in 2015.
"These kinds of videos suggest, likewise, that the vocal tract more broadly may have the potential to produce speech-like sounds.
"I know of a video with a husky that says, 'I love you.' "
Pity our animal cousins. Their throats and mouths are the wrong shape to form the many vowel sounds that slip between our consonants like grease between gears — sorting a clog of fricatives and plosives into the works of Dante or Celine Dion.
At least, that was the common thinking when Dion was little. It dates back to a study on a dead rhesus monkey in 1969, said Tom Sawallis, a linguist at the University of Alabama.
"You've got to have contrasting vowels to have the vocabulary, and you've got to have vocabulary to have syntax," he said. And so on, all the way up to carpool karaoke.
"If you can't have contrasting vowel sounds until 400,000 years ago, then you can't have had language until then, at the very earliest," Sawallis said.
And so supposedly, the most disruptive innovation in evolution developed in a few hundred thousand years, in the human body alone.
That thinking is changing.
Last year, a team of researchers took X-ray videos of macaque monkeys' throats — live ones, not dead, as they went about their days.
The footage showed that the animals' vocal tracts were not so ill-suited for speech after all. The macaques appeared capable of making nearly every human vowel sound.
Enter a team led by French researchers — Sawallis the lone American among them — who published their own study in Plos One this month after recording audio of monkeys.
"We found the actual vowel qualities that are used in human language," Sawallis said.
They listened in on baboons, not macaques. But they picked up five vowel sounds that fit neatly into the same phonetic alphabet you can find in a classroom dictionary: A as in an American apple, O as in the British "bought."
"Monkeys are capable of talking," Sawallis said. At least, physically.
That would mean that what we now call language has been baking in the genetic code since the days of our last common ancestor with baboons — some 25 million years ago — if not longer.
"All the theories need re-examination," Sawallis said.
Even he would not go as far as Scott Moisik, a researcher with the Max Planck Institute for Psycholinguistics, whose studies have thus far focused on human voices — but whose ears pricked up when he heard what Sawallis's group found.
"When I hear a cat on YouTube produce a vocalization that very much sounds like 'oh long Johnson,' " he said in an email to the Associated Press, he wonders how far back down the evolutionary tree our path to speech began.
He told The Washington Post that he hears in house pets not just vowel qualities but intonation, syllabicity, rhythm "and, amazingly (especially in the case of cats) some consonantal strictures."
"However these animals do it . . . one cannot deny there are properties in their vocalization that are approximating human speech," Moisik wrote. "Someone needs to take the time and really do the analysis."
Sawallis agrees on the last point, though he's skeptical about the talking pet stuff.
"We as humans are really, really good at pattern detection. We overrecognize in a variety of ways," he said. "But I don't have to certify that 'Oh Long Johnson' does do something interesting to know we should investigate these other animals."
So, too, thinks Marcus Perlman, a cognitive scientist who also researches with the Max Planck Institute.
"The studies with monkeys show it's not particular to humans," he said. "This might extend more broadly."
He was curious about Mishka, the viral husky that appeared to whine "I love you."
Perlman's research has focused instead on great apes — a closer relative to humans than monkeys, though no less disruptive to scientific consensus.
Whatever sounds make up a vocal repertoire — vowels, consonants, grunts or basic barks — humans were once thought the only primate able to control their voices to any significant extent.
Other animals were thought to make sounds like you yelp when you touch an iron: as pure reflex.
But in a 2015 study, Perlman and a team of researchers documented how a 280-pound gorilla named Koko had been trained to cough, blow into a recorder and make several other noises at will.
His team's study was followed up last year, when an orangutan named Rocky at the Indianapolis Zoo was trained to control something approximating a voice.
"He was able to listen to a human make a vocalization and able to match the frequency of that," said Rob Shumaker, who is the zoo's executive vice president and co-authored the study. "Prior to this study with Rocky, most of the conversation was saying this is a uniquely human event."
Put that together with the Koko study, and the baboons and macaques, and "we are starting to put the puzzle together."
And might a talking house cat fit somewhere in the evolutionary jigsaw?
In a word, Shumaker said, no.
The YouTube videos are interesting, he agreed. But he pointed to parrots — long known to mimic human language perfectly but using body parts so alien that they say little about our course of evolution.
"I'm very skeptical that what dogs and cats are doing is really going to inform us about the origins of human language," Shumaker said.