The Language of Whales: Decoding Cetacean Communication

A persuasive, visually immersive 12-slide deck exploring the scientific breakthrough in decoding whale communication — from sperm whale phonetic alphabets to AI-powered translation efforts — and why it demands we rethink language, intelligence, and legal rights.

The Language of Whales

For decades we listened. Now, for the first time, we are beginning to understand. Inside the scientific revolution decoding cetacean communication — and what it means for life on Earth.

What We'll Explore Today

A Signal in the Deep: The Discovery

The Science of Whale Song

Culture, Society, and Learning

Project CETI: Decoding the Language

Why This Changes Everything

We Thought It Was Morse Code. It's Far More.

For decades, scientists treated whale clicks as simple binary signals — like dots and dashes

New research reveals combinatorial structure: whales combine sounds into patterned sequences, much like human phonemes form words

Sperm whale codas contain 3–40 broadband clicks arranged in culturally distinct patterns

This is not signaling. This is something closer to syntax.

Sources: 1, 2

A Phonetic Alphabet Hidden in the Abyss

Project CETI identified structures analogous to vowels and consonants in sperm whale vocalizations

Whales from different clans use distinct 'dialects' — culturally transmitted, not genetic

These dialects persist across generations within matrilineal family groups

The communication system is socially learned, not hardwired.

Sources: 1, 2, 3

04

The Science of Whale Song

Humpback Songs Obey the Same Statistical Laws as Human Language

Researchers applied infant speech segmentation algorithms to 8 years of humpback recordings

Whale song exhibits a power-law frequency distribution — a hallmark of all human languages

This statistical structure is thought to facilitate learning and faithful cultural transmission across generations

Two evolutionarily distant species converge on the same acoustic architecture — pointing to universal principles of learned communication.

Sources: 4, 5

This finding, published in Science, is breathtaking in its implications. When human infants learn language, they rely on statistical regularities — certain sound patterns appear much more frequently than others, and this predictability helps them segment the continuous stream of speech into words. That frequency distribution follows what's called a power law — and it's present in every known human language. When the team applied those same infant speech-segmentation algorithms to humpback whale songs, they found the exact same statistical structure. Not similar. The same. This convergence — in two species separated by tens of millions of years of evolution — strongly suggests that if you have a communication system that must be learned and culturally transmitted, certain statistical properties may be universal.

Culture Runs Deeper Than Instinct

Cetaceans live in matrilineal societies where knowledge, hunting techniques, and songs are passed down socially

Humpback songs evolve rapidly: themes shift and recombine within a single season, yet remain recognizable across years

Norwegian feeding-ground studies confirm directional, non-random evolution of song structure over consecutive years

This is not random variation. It is cumulative cultural evolution.

Sources: 3, 5

07

Project CETI

The Road to Translation

1960s

2000s

2010s

2020

2024

2025

Roger Payne records humpback songs; public awakening

Digital acoustic monitoring enables large-scale cetacean data collection

Machine learning applied to bioacoustics for the first time

Project CETI launches — linguistics, AI, robotics, and marine biology unite

Combinatorial structure in sperm whale codas confirmed (Nature Communications)

Phonetic alphabet evidence published; legal rights discourse begins

Sources: 1, 2, 7, 8, 9, 10

Let me take you through how we got here. In the 1960s, biologist Roger Payne dropped the needle on a vinyl record called 'Songs of the Humpback Whale' and the world listened for the first time — it became the best-selling nature recording in history and helped launch the Save the Whales movement. Fast forward to the 2000s, and digital acoustic monitoring meant we could finally gather enough data to do serious science. By the 2010s, machine learning entered bioacoustics, and suddenly patterns invisible to the human ear became visible to the algorithm. Then in 2020, Project CETI was born — an unprecedented collaboration of linguists, roboticists, AI researchers, and marine biologists. The last two years have delivered the combinatorial structure paper in Nature Communications, and in 2025, the first evidence of a phonetic alphabet — alongside the first serious legal scholarship asking what rights these beings might hold.

AI Is the Rosetta Stone of the Deep

Project CETI deploys robotic sensor arrays, machine learning, and computational linguistics to map whale communication

AI models trained on millions of coda sequences can now predict the next click in a sequence — the same capability that powers large language models

Video and audio recordings now capture full behavioral context: births, social bonding, group decisions

The goal: not just to decode, but to understand meaning in context.

Sources: 7, 9, 10

“

Is there anything uniquely human about language, or is it just a continuum? What does that mean for the law?

Dr. Gašper Beguš, Linguistics Lead, Project CETI — UC Berkeley

Sources: 1, 10

We Are Not Alone in Language

The question is no longer whether whales communicate. It's whether we have the wisdom to listen.

References

[1] UC Berkeley and Project CETI study shows sperm whales communicate in ways similar to humans | Letters & Science — ls.berkeley.edu [2] Contextual and combinatorial structure in sperm whale vocalisations | Nature Communications — nature.com [3] Life in the pod: the social lives of whales | Natural History Museum — nhm.ac.uk [4] Whale song shows language-like statistical structure - Science — science.org [5] Frontiers | Changes in humpback whale song structure and complexity reveal a rapid evolution on a feeding ground in Northern Norway — frontiersin.org [7] Groundbreaking effort launched to decode whale language — nationalgeographic.com [8] Toward understanding the communication in sperm whales — pmc.ncbi.nlm.nih.gov [9] Project CETI - National Geographic Society — nationalgeographic.org [10] AI-enabled decoding of whale communication could bolster animal rights, César Rodríguez-Garavito argues | NYU School of Law — law.nyu.edu [PDF] Analysis of Humpback Whale Songs using Information Theory ... — srs.amsi.org.au

More AI presentation examples · Create your own with Sliderest