​A Thousand Dollar Prize Awaits Whoever Can Speak Without Words

Want to make a thousand bucks? How good are you at grunting? All you have to do is make a sound—but not a word—for each word on a list, that some stranger will then listen to on the internet, and have to guess which word your grunt meant. Whoever's grunts get the most matches, wins.

It may sound silly, especially as I've described it, and maybe it is. But it's silly in service of a grand ambition. It's an experiment that aims to turn back the clock of evolution, strip off thousands of years of culture as much as possible, and try to shed light on how one humanity's most characteristic traits—Language with a big L—came to be.

But, also it's a chance to win a thousand bucks for what is literally gibberish.

Here are the rules: Competing teams will make up vocalizations for "30 Paleolithic-relevant meanings," ranging from actions like "eat" or "cook" and things like "tiger" or "water" and properties like "dull" or "sharp" or "many." The vocalizations can't be words, obviously, and they can't be "emblems" of speech, like using "booo" for bad, or conventional imitations of their subject—no hissing to mean "snake."

Once all of the sounds are collected, people from Amazon Mechanical Turk will be paid to listen to the vocalizations and pick out what they're supposed to mean. Whoever's vocalizations score the highest wins the grand "Saussure Prize."

Everything about the proceedings seem straightforward enough, except why someone would ever want to do this, and who the hell Saussure is. So I reached out to the creator for some answers.

Marcus Perlman isn't some weirdo with a caveman fetish—actually, full disclosure, I didn't make him deny that he was, but, at any rate, that's not why he set up the VIC. Perlman is a post-doc in the University of Wisconsin's department of psychology, and he told that the point of the VIC is "to understand how languages are created—how linguistic conventions (e.g. words) come to be established."

Studies like this are a core mission of the Lupyan Lab at Wisconsin, which examines the interaction of language and other cognitive processes. Their lab philosophy is to find questions that people don't think can be tested empirically, and then to test them empirically. The question of how humans developed language falls into that category.

"Language doesn't leave fossils," Perlman told me via email, "it appears to be very different from the communication systems of other animals; historical linguistics cannot reconstruct languages beyond a few thousand years (compared to 10s or 100s of thousands of years); we've never witnessed the creation of a spoken language, and it is obviously unethical to conduct realistic experiments to observe their creation."

"Even the most obscure of languages," he adds, "are extremely ancient beyond any current ability to detect any signal of the primordial language."

As a result of all these obstacles, there isn't a single prevailing theory on where language came from. "In fact, there is considerable disagreement even concerning the most fundamental questions, such as whether language evolved from vocalizations or gestures," Perlman said. "The gestures vs. vocalizations question is where the Vocal Iconicity Challenge comes in."

Language today generally is not iconic—the words don't need to reflect the thing they signify. Because of this, it has become a sort of dogma in linguistic circles to believe that the relationship between a word's form and its meaning is arbitrary. "In a seminal language evolution paper published in 1990, the cognitive scientists Steven Pinker and Paul Bloom write that arbitrariness in language is 'most obvious in the choice of individual words: there is no reason for you to call a dog dog rather than cat except for the fact that everyone else is doing it,'" Perlman told me.

The man popularly credited for coming up with this theory is the late 19th- and early 20th-century linguist Ferdinand de Saussure. Naming the prize of the VIC after Saussure was Perlman's "attempt at humor," he said, because this principle is what the VIC is supposed to test.

"We're focusing on a very particular aspect of the language evolution process—can humans generate novel vocalizations that they can then conventionalize into words? Or, as a major language evolution theory alternatively proposes, must humans first generate novel gestures to communicate a meaning, and then in turn use these gestures to bootstrap arbitrary vocalizations," Perlman said. "A major theory of language evolution proposes that, while people very naturally produce novel meaningful(i.e. "iconic") gestures to communicate different meanings, they cannot create novel meaningful vocalizations. We are testing this claim."

Obviously the VIC isn't going to definitively answer that question either way, but if people are able to create a sort of ad hoc language, that's going to give credence to the idea that words, or proto-words were iconic, rather than arbitrary. And it's going to give Perlman something weird to listen to.

For whatever its worth, Perlman assured me that he expects "that the vocalizations that contestants submit will be more sophisticated than 'grunting.'" I'm sure that's true, but that doesn't mean there won't be tons of great grunts coming from this project. And also, naturally, a thousand bucks.