Cambridge, MA, May 7, 2024 — Project CETI (Cetacean Translation Initiative) today published research in Nature Communications led by members of MIT’s Computer Science & Artificial Intelligence Laboratory (CSAIL), for the first time proposing the “sperm whale phonetic alphabet” and revealing that sperm whale communication has both contextual and combinatorial structure not previously observed in whale communication.
Sperm whales communicate primarily using sequences of short bursts of clicks with varying inter-click intervals, known as codas. The paper reports previously undescribed variations in coda structure that are sensitive to the conversational context in which they occur. This study shows that coda types are not arbitrary, but rather that they form a newly discovered combinatorial coding system in which the musical concepts of rubato and ornamentation combine with two categorical, context-independent features known as rhythm and tempo, by analogy to musical terminology.
Rhythm, tempo, rubato and ornamentation can be freely combined. This gives rise to a large inventory of distinguishable codas — a “sperm whale phonetic alphabet” that makes it possible to systematically explain observed variability in coda structure.
“While the communicative function of many codas remains unknown, these results show that the sperm whale communication system may represent a large space of possible meanings,” said Dr. Daniela Rus, Director of MIT Computer Science & Artificial Intelligence Laboratory and a member of CETI’s machine learning and robotics teams.
“Investigating a dataset collected across over a decade of nearly 9,000 codas from the sperm whale families of the Eastern Caribbean clan, our results show that these whales have a more complex combinatorial communication system that demonstrates rubato and ornamentation, in which whales make sub-second adjustments to match one another as they converse and add extra clicks to known coda types depending on the context within their conversations. Despite their dramatically divergent evolutionary lineage, sperm whales have aspects within their communication system typically reserved for humans,” said Dr. Shane Gero, Biology Lead of Project CETI, Scientist in Residence at Carleton University and Founder of the Dominica Sperm Whale Project.
“This paper shows that sperm whale communication has some of the same structural features as the most sophisticated communication systems in the animal kingdom. We’re excited to start studying how it’s used to convey meaning,” said Jacob Andreas, Assistant Professor of Electrical Engineering and Computer Science at MIT and member of the Project CETI Machine Learning Team.
”Communicative or not—it is all in the context! It is only in the wider communicative context that the gradual change in calls and the imitation of the changes becomes clear. It is also only then we find some clicks act like a suffix to an unchanging rhythmic beat and not all clicks should be grouped together,” said Pratyusha Sharma, PhD student at MIT’s CSAIL and member of Project CETI Machine Learning Team.
“This discovery marks a profound moment in advancing our understanding of sperm whales,” said Dr. David Gruber, Founder and Lead of Project CETI. “It opens up the possibility that sperm whales have an incredibly complex and nuanced communication system –and inspires us to continue on our whale listening journey.”
CETI’s field research on sperm whale vocalizations may investigate if rhythm, tempo, ornamentation, and rubato encode whales’ communicative intents. Additionally, it’s possible that sperm whale communication might provide the first example of the linguistic concept “duality of patterning,” which refers to a set of individually meaningless elements that can be combined to form larger, meaningful units outside of human language.
The paper “Contextual and Combinatorial Structure in Sperm Whale Vocalisations” is written by:
- Pratyusha Sharma, Project CETI Machine Learning Team, PhD student, MIT’s CSAIL
- Shane Gero, Project CETI Biology Lead, Scientist in Residence at Carleton University, Founder of the Dominica Sperm Whale Project and National Geographic Explorer
- Roger Payne, Project CETI Former Advisor, Whale Biologist, Conservationist and MacArthur Fellow
- David Gruber, Project CETI Founder and Lead, Distinguished Professor of Biology, Baruch College, City University of New York and National Geographic Explorer
- Daniela Rus, Project CETI Machine Learning and Robotics Team, Director, MIT’s CSAIL
- Antonio Torralba, Project CETI Machine Learning Team, Professor of Electrical Engineering and Computer Science at MIT
- Jacob Andreas, Project CETI Machine Learning Team, Assistant Professor of Electrical Engineering and Computer Science, MIT
Dr. Roger Payne, who discovered that humpback whales sing in the famous Science article, Songs of Humpback Whales in 1971, is a participant of this study. Dr. Payne’s research sparked the “Save the Whales Movement” — one of the most successful conservation initiatives in history. Project CETI aims to build upon Dr. Payne’s legacy by using technology and scientific discovery to bring humans closer to nature. After the paper was submitted for peer review, Roger published an article in TIME five days before his death, highlighting his hopes for CETI’s research. Project CETI is proud to build upon this legacy, demonstrating how technology can be used to amplify the magic of our natural world and bring us closer to nature.
The paper was published in Nature Communications on May 7, 2024. More information, including a copy of the paper, can be found online: https://www.nature.com/articles/s41467-024-47221-8.
This analysis was funded by Project CETI via grants from Dalio Philanthropies and Ocean X; Sea Grape Foundation; Virgin Unite, Rosamund Zander/Hansjorg Wyss, and Chris Anderson/Jacqueline Novogratz through The Audacious Project: a collaborative funding initiative housed at TED. Further funding was provided by the J.H. and E.V. Wade Fund at MIT.
Project CETI thanks the Chief Fisheries Officers and the Dominica Fisheries Division officers for research permits and their collaboration in data collection; all the crews of R/V Balaena and The Dominica Sperm Whale Project team, and their funders, for data collection, curation, and annotation of the dataset; as well as Dive Dominica, Al Dive, and W.E.T. Dominica for logistical support while in Dominica.
###
About Project CETI
Project CETI (Cetacean Translation Initiative) is a nonprofit organization applying advanced machine learning and state-of-the-art robotics to listen to and translate the communication of sperm whales in the Eastern Caribbean off the coast of the island of Dominica. Its science team comprises leading artificial intelligence, natural language processing and complex systems experts, cryptographers, linguists, marine biologists, roboticists, engineers, and underwater acousticians in partnership between over 15 academic institutions and companies across six countries. In 2022, it published in iScience its scientific roadmap for advancing the understanding of sperm whale communication. This peer-reviewed academic paper outlines the key elements required for the collection and processing of massive datasets, detecting basic communication units and language-like higher-level structures, and validating models through interactive playback experiments.
For media inquiries please contact McPherson Strategies at ceti@mcpstrategies.com.