top of page

Article: Are We a Step Closer to Understanding What Whales Say?

Photo Source: Wikimedia Commons

On July 4, 2024, a whale-watching cruise out of Gloucester, MA, enjoyed this once in twenty years rare sighting of a blue whale in the waters off beautiful Cape Ann. ( Usually only found far offshore and closer to the Arctic, this endangered blue whale surprised onlookers as it foraged for the bountiful krill in the cool waters of the Atlantic Ocean. Such an unusual sighting turned my thoughts to research going on to understand the social life and languages of whales. Researchers using artificial intelligence have made a breakthrough in understanding sperm whale communication.


Marine mammal communications, such as among whales or dolphins, have long fascinated researchers seeking to understand the evolution of language and intelligence. While the human language possesses great complexity, recent studies suggest that some animal communication systems may be more sophisticated than previously thought, particularly among highly social species like dolphins and sperm whales.


Dolphins, for instance, use sequences of clicks and whistles to communicate. Experiments have shown that dolphins produce more vocalizations when cooperating to solve puzzles, suggesting their sounds convey meaningful information. In research led by Holi Eskelinen, the scientists found that dolphins produced more chatter while cooperating to solve the tasty puzzle. ( When the researchers put dolphin treats like fish in a clear plastic tube with caps on each end and ropes attached to the caps, dolphins figured out that they needed to grab a rope and pull it to release the treats. Although the content of their communication remains a mystery, the researchers concluded that the communication had different qualities compared to social chatter or communication when foraging. 


Artificial intelligence has made great strides in modeling human language. Large language models like ChatGPT can now translate between languages with remarkable accuracy. This success has inspired researchers to apply similar AI techniques to animal communication. In 2020, an ambitious research program called Project CETI, which stands for Cetacean Translation Initiative, began with the goal of translating the language of sperm whales. The organizers, including Roger Payne, describe CETI as “a transdisciplinary research initiative bringing together leading technologists, roboticists, cryptographers, linguists, and marine scientists to study the communication of the world’s largest toothed predator: the sperm whale.” ( The sperm whale, according to, can grow up to fifty-two feet in length and weigh up to 90,000 pounds. They hunt down to 10,000 feet deep in the ocean for squid, sharks, and other fish. Because baby sperm whales cannot dive so deep, females form daycare pods to look after the calves near the surface while other whales dive for food. Sperm whales have been particularly interesting to researchers due to their large brains, complex social structures, and distinctive vocalizations called “codas” -- rhythmic sequences of clicks used for communication.


However, new research published in 2024 in the journal Nature Communication reveals that sperm whale communication exhibits significantly more complexity than previously understood. The study, led by Pratyusha Sharma and colleagues at MIT and Project CETI in an article titled “Contextual and combinatorial structure in sperm whale vocalisations,” analyzed over 8,700 codas recorded from sperm whales in the Caribbean. (


The researchers discovered that sperm whale codas have a combinatorial structure built from four key features:


Rhythm: The relative timing of clicks within a coda


Tempo: The overall speed or duration of the coda


Rubato: Gradual changes in tempo across sequences of codas


Ornamentation: The addition of extra clicks to standard coda patterns


Importantly, these features can be combined independently, allowing whales to produce a vast array of distinct vocalizations from a limited set of elements - similar to how human languages combine phonemes or sounds to form words. Humans have 44 phonemes or sounds we make to form words, such as the letters of the alphabet and combined letters such as ch and th. How we say words and their context make our communication complex and varied.


The study found eighteen distinct rhythm types and five tempo categories. When combined with optional rubato and ornamentation, this system allows for over 140 coda variations - far more than the 21 coda types previously identified in this whale population. The researchers found that sperm whales adjust their vocalizations based on their conversational situation. Whales often match the tempo of codas produced by other whales they are communicating with. They also use ornamentation (extra clicks) in specific positions, such as at the beginning or end of coda sequences, possibly to signal changes in the conversation. While the full meaning of different codas remains unknown, this system theoretically allows whales to convey much more complex information than a set of fixed call types.


These findings open exciting possibilities for future research. Could sperm whales have a communication system approaching the complexity of human language, with the ability to convey abstract concepts or discuss past and future events? However, researchers caution against anthropomorphizing or oversimplifying animal communication. Studies of human communication have shown that only about 7% of meaning comes from the words themselves, with the majority conveyed through nonverbal cues and vocal tone. Animal communication likely relies heavily on non-vocal elements that may be difficult for humans to perceive or interpret.


As artificial intelligence and other technologies advance, researchers hope to gain deeper insights into the communication systems of whales, dolphins, and other highly intelligent animals. Understanding animal communication could have profound implications for conservation, animal welfare, and our conception of intelligence and consciousness in different species. While we may never fully decipher whale language, continued research promises to reveal new wonders about the rich inner lives of our planet’s largest-brained creatures.

 Dr. Smith’s career in scientific and information research spans the areas of bioinformatics, artificial intelligence, toxicology, and chemistry. He has published a number of peer-reviewed scientific papers. He has worked over the past seventeen years developing advanced analytics, machine learning, and knowledge management tools to enable research and support high-level decision making. Tim completed his Ph.D. in Toxicology at Cornell University and a Bachelor of Science in chemistry from the University of Washington.

You can buy his book on Amazon in paperback and in kindle format here.


bottom of page