What Can a Bonobo Teach Us About Language?

Phenomenal article telling the story of bonobos language. Are they human or not?

To some scientists, Kanzi’s intellectual feats demonstrated clearly that language was not unique to human beings. But others were unimpressed. “In my mind this kind of research is more analogous to the bears in the Moscow circus who are trained to ride unicycles,” said the Harvard psychologist Steven Pinker. To him, the fact that Kanzi had learned to produce elements of human communication didn’t imply that he had the capacity for language. Thomas Sebeok, a prominent linguist who organized a conference in 1980 that helped squelch public funding for animal language research, had a similar take. “It has nothing to do with language, and nothing to do with words,” he said, when asked to comment on Savage-Rumbaugh’s work. “It has to do with communication.”

The controversy masked an uncomfortable truth: No one agreed on what the difference between language and communication actually was . The distinction goes back to Aristotle. While animals could exchange information about what they felt, he wrote, only humans could articulate what was just and unjust, and this made their vocalizations “speech.” In the 1600s, the philosopher René Descartes echoed this idea: While animals jabbered nonsensically, he wrote, God had gifted human beings with souls, and with souls language and consciousness. In the modern era, the influential linguist Noam Chomsky theorized that human beings possess a unique “language organ” in the brain. While human languages might sound and look different from one another, Chomsky wrote in the 1960s, all of them are united by universal rules that no other animal communication system shares. According to Chomsky’s early work, this set of rules distinguishes the sounds and gestures we make when we talk from the dances of bees, the twittering of birds and the spectral keening of whales. It’s the magic ingredient that makes our languages uniquely capable of reflecting reality.


I can do that. They’re related inversely. Judging from the internet, language is a tool designed to prevent communication.


The question seems to be different n the article… its more like -
Are humans made by human socio-cultural norms?

Hi @swamidass,

As someone whose Ph.D. is on the subject of animal minds, I have to say I was underwhelmed by the Smithsonian magazine article, What can bonobos teach us about the nature of language?

First, the claim that scientists cannot agree on what the difference between language and communication actually is, is a misleading one. May I refer you to the following article:

Alex, B. (2018). “Could Neanderthals speak? The Ongoing Debate over Neanderthal language”. Discover , November 6, 2018. A brief excerpt will suffice:

“Without straying too far into academic debates over the nature of language, let’s just say there are broad and narrow theories when it comes to what actually constitutes language.

“A broad view defines language as a communication system in which arbitrary symbols (usually sounds) hold specific meanings, but are not fixed or finite. Words can be invented, learned, altered and combined to convey anything you can think.

Narrow definitions focus on syntax and recursion, structural properties shared by all human languages today. These both refer broadly to the set of rules that guides how statements can be formulated in any given language, and they are thought to be hardwired into our brains. By this view language is ‘a computational cognitive mechanism that has hierarchical syntactic structure at its core,’ in the words of biologist Johan Bolhuis and colleagues.”

The point I want to make is that Kanzi’s attempts at communication don’t even satisfy the broad criterion of language, let alone the narrow one. Chimpanzee vocalizations cannot be combined to make an infinite number of sentences, for instance. Nor can chimpanzee vocalizations (or lexigrams, for that matter) be “invented, learned, altered and combined to convey anything you can think.”

Allow me to quote from If a lion could talk: animal intelligence and the evolution of consciousness (The Free Press, 1998) by Stephen Budiansky, a former Washington editor of Nature, whose book was highly praised by Sir John Maddox (Editor Emeritus of Nature) as well as by Elizabeth Marshall Thomas, author of The Hidden Life of Dogs.

Kanzi’s learning by observation as [his mother] Matata pushed lexigrams and received various rewards is one of the few thoroughly documented instances of observational learning in animals. It remains far from clear, however, that any of this use of symbols equates to an understanding of symbols… There is indeed a semantic trick in even calling one of those lexigrams “please.” Why “please”? All that key does is initialize the computer. Does the ape understand the concept of “please”? (p. 154)

Savage-Rumbaugh reports that 96 percent of Kanzi’s utterances take the form of demands for food, tickles, or other activities. As we shall see, this fact alone implies something fundamentally different about what Kanzi is understanding and what a human child is. (p. 156)

Children will say “red” or “plane” or “cat” in circumstances where they show no interest at all in obtaining the object referred to. Uttering the name is a goal in itself. This behavior appears spontaneously at around nine to thirteen months. (p. 157)

It is the fundamental difference between using symbols and understanding symbols that constitutes the discontinuity between animals and humans, and which leads to the manifest and huge gap between the rote demands of language-trained apes and the conceptual flights of humans. Savage-Rumbaugh has argued that in terms of quantitative acquisition of “words” and combination of those words into sentences, Kanzi equals the abilities of a two-and-a-half-year-old human. But the qualitative difference in the way Kanzi employs words from the way even a one-year-old child does is vast, and is what allows a child to keep on expanding its grasp and use of language. “If a child did exactly what the best chimpanzee did,” says [critic Herbert] Terrace, “the child would be thought of as disturbed.” (pp. 159-160)

Finally, I note that neither of the authors of the Smithsonian article has a science background. One is a photographer, and the other is currently studying for a Ph.D. in comparative literature. Although she has authored a novel, this is her first article for Smithsonian magazine. May I courteously suggest that caution is called for, when assessing the credibility of such an article.


@vjtorley, it is scientific journalism, not the final say on these matters. Rather, it is an opportunity to enjoy the story and fill in the details.

I would entirely agree, as would most linguists.

1 Like

I did some work for Dr. Sebeok’s department back in that very year of 1980. One of the many interesting aspects of his semiotics research was helping to develop the cartoon-like nuclear waste warning messages that are meant to alert humans and perhaps other beings for thousands of years into the future after the English language is long forgotten:

Even more interesting was Sebeok’s proposal to create an atomic priesthood which would use the power of religion to preserve the knowledge of the radioactive wastes’ dangers through “sacred texts” and rituals.

Sebeok and the other Hungarians in that department were some of the most interesting people I ever worked with. They had fascinating stories from WWII and also the day that Soviet tanks crushed Hungary’s rebellion of 1956.

1 Like