Humpback Whale Song Shares a Key Pattern With Human Language That Might Make It Easier for the Animals to Learn
Despite humans and whales being separated by millions of years of evolution, our vocalizations follow the same principle outlined in Zipf’s law
/https://tf-cmsv2-smithsonianmag-media.s3.amazonaws.com/accounts/headshot/Margherita_Bassi.png)
All human languages follow the same pattern: The most common word is used twice as often as the second most common word, three times as often as the third most common word, four times as often as the fourth most common word… and so on.
This statistical rule is called Zipf’s law, and now, an interdisciplinary team of scientists has revealed that humpback whale songs follow the same pattern. In a study published Thursday in the journal Science, researchers suggest this rule might have developed because it makes culturally transmitted communication such as human language—and the songs of humpback whales—easier to learn.
“It strengthens the view that we should be thinking about human language not as a completely different phenomenon from other communication systems but instead think about what it shares with them,” Inbal Arnon, a psychologist at the Hebrew University of Jerusalem and a co-author of the study, tells Scientific American’s Cody Cottier.
How to search for the pattern in humpback whale song, however, wasn’t immediately clear. First, the researchers had to identify the units within the song that would be equivalent to human words. To achieve this, they were inspired by the way babies learn language.
“Speech is continuous, and there are no pauses between words, so infants have to discover word boundaries,” Arnon explains to New Scientist’s James Woodford. “To do this, they use low-level statistical information: Specifically, sounds are more likely to occur together if they are part of the same word. Infants use these dips in the probability that one sound follows another to discover word boundaries.”
For example, children could ultimately identify the distinct words in the phrase “after dinner” by noting that the syllables “ter” and “din” do not usually go together. “If whale song has a similar statistical structure,” Arnon adds to New Scientist, “these cues should be useful for segmenting it as well.”
The researchers used this method to analyze the distribution of these segments in eight years of humpback whale song data collected in New Caledonia in the South Pacific. They plotted the frequency of sounds and word-like sequences—including syllables they describe as “grunt,” “descending high squeak” and “ascending moan”—and realized that humpback whale songs also follow Zipf’s law.
“I’ll never forget the moment that graph appeared, looking just like the one we know so well from human language,” says study co-author Simon Kirby, a cognitive scientist at the University of Edinburgh, to New Scientist. “This made us realize that we’d uncovered a deep commonality between these two species, separated by tens of millions of years of evolution.”
Why would two species both develop communication that follows Zipf’s law? The researchers suggest it might be because it makes that form of communication—whether it’s language or song—easier to pass on to the next generations.
The presence of Zipf’s law in humpback whale songs “suggests their communicative behavior has culturally evolved to become easier for novices to learn, echoing how the structure of our languages seems to have evolved to be more learnable by human infants,” Andrew Whiten, a primatologist at the University of St. Andrews in Scotland who was not involved in the study but co-authored an accompanying perspective on it, tells Tim Vernimmen for National Geographic.
With future research, “we should find these statistical properties in any culturally transmitted system of sequential signaling,” Arnon tells the New York Times’ Emily Anthes. “So, we have bats to look at, we have songbirds to look at, we have elephants, maybe, to look at.”
In a related paper published Wednesday in the journal Science, Mason Youngblood, a researcher at Stony Brook University, found two other shared features between human language and the vocal communication methods of humpback whales—as well as in several other species of whales and dolphins. One is Menzerath’s law, which says that lengthier sequences (such as a long sentence) usually have smaller components (such as shorter words) when compared to shorter sequences. The other is the brevity law, which holds that more common words tend to be shorter.
Previously, scientists have discovered related patterns in the language of other animals. For instance, jackass penguins—named for their vocalization that resembles a “donkey in distress”—were found in 2020 to follow the brevity law and Menzerath’s law, too.
However, the researchers from the study inspired by babies’ learning emphasized that whale song is not equivalent to human language, namely because it lacks expressive meaning, according to a statement from the University of St. Andrews. On the other hand, humpback whale song could be comparable to human music, which also lacks semantic meaning while still adhering to Zipf’s law.
But that doesn’t make the study less exciting. The new work nevertheless challenges “long held assumptions about the uniqueness of human language, uncovering deep commonalities between evolutionarily distant species,” Kirby says in the statement.
