Whale songs are very similar to human language, study finds

Language has long been seen as a defining trait of human intelligence. Unlike the communication systems of other species, human language consists of predictable, recurring elements that follow statistical patterns. These structures aid learning and transmission across generations.

A recent study suggests that these same features may exist in the complex vocal displays of humpback whales.

Researchers have discovered that whale songs share statistical properties found in human speech. By applying methods inspired by how infants segment language, they uncovered a pattern in whale vocalization that had previously gone undetected.

Findings, published in the journal Science, suggest a deep commonality between two species separated by millions of years of evolution, both relying on cultural transmission to shape their communication.

Male humpback whales produce long, structured songs made up of smaller units. These vocalizations follow a hierarchical pattern. Individual sound elements combine into phrases, repeated to form themes. Multiple themes create a song, and several songs make up a session.

The entire male population of a region tends to conform to a common song structure. Over time, the song gradually evolves through subtle changes in themes. Occasionally, a song revolution occurs, where a new arrangement rapidly replaces the existing one. These revolutions have been observed spreading across the South Pacific, demonstrating how cultural learning drives song evolution.

Scientists have struggled to understand how whales learn these complex songs. Unlike birds, whales cannot be studied in controlled environments. However, observations of hybrid songs—where a whale is caught in the process of switching melodies—have provided valuable insight.

Analyses suggest that whales learn by inserting new song segments into familiar structures, a process similar to how human infants acquire language.

To test whether whale songs share properties with human language, researchers analyzed eight years of recordings from a single whale population. They applied a segmentation method modeled after how babies detect words in speech. This method identifies points where the probability of one sound following another suddenly drops—a clue infants use to recognize word boundaries.

Results showed that whale songs contain statistically coherent sequences, much like words in human speech. Moreover, the distribution of these sequences followed a power-law pattern, a key feature of language. This statistical structure had never before been observed in non-human communication.

While whale song does not carry semantic meaning, its structure may resemble human music, which also follows predictable statistical patterns. These findings suggest that the way communication systems evolve may be shaped by cultural transmission rather than genetic inheritance alone.

Dr. Ellen Garland from the University of St Andrews described the discovery as unexpected yet significant:

Dr. Inbal Arnon from the Hebrew University highlighted the role of learning in shaping communication systems:

Professor Simon Kirby from the University of Edinburgh emphasized the broader impact of these findings:

This research challenges the long-standing assumption that human language is entirely unique. It suggests that the process of cultural transmission—rather than biological evolution alone—plays a crucial role in shaping complex communication systems.

Whether in human speech or whale song, the ability to learn, adapt, and pass down vocal patterns may be a fundamental trait of intelligent species.

Note: Materials provided above by The Brighter Side of News. Content may be edited for style and length.

Like these kind of feel good stories? Get The Brighter Side of News' newsletter.