Whale Alphabet Discovered In Landmark Scientific Experiment

By Charlene Badasie | Published

Researchers have discovered a new level of complexity in the language of sperm whales. Through analysis of thousands of recorded sequences using artificial intelligence, scientists have uncovered a wealth of information about the click sounds (or whale alphabet) created by air being forced through an organ in the creatures’ heads called the spermaceti.

The Whale Alphabet

The study published in the journal Nature Communications sheds light on the remarkable intricacies of the sperm whale alphabet.

Led by a team including Dr. Daniela Rus, director of the Computer Science and Artificial Intelligence Laboratory (CSAIL) at MIT, the research marks a significant leap forward in understanding the communication systems of these majestic marine mammals.

At the heart of the whale alphabet study are the click sequences, known as codas. They are characterized by variations in tempo, rhythm, and length, forming the basis of the whales’ acoustic repertoire. Through their analysis, researchers have found that sperm whales possess the ability to combine these click patterns in various ways to convey information to one another.

A Phonetic Alphabet

While the precise meaning behind these clicks remains elusive to humans, the identification of distinct patterns represents a crucial step toward deciphering the whale alphabet.

The researchers have coined the term “phonetic alphabet” to describe the sound combinations observed in sperm whale communication, drawing parallels to the phonetic sounds in human speech.

However, Dr. Luke Rendell, a communication researcher at the University of St. Andrews, cautions against interpreting these findings through a strictly human-centric lens. He emphasizes that while the whale alphabet study provides valuable insights, it is essential to avoid imposing anthropocentric perspectives on the complexities of whale communication.

Machine Learning

 whale alphabet

“The way the tempo variation is used is completely different to how, say, we use elements of an alphabet to construct linguistic expression,” Rendell said. “There’s no evidence of that, and it’s not a super helpful interpretation because it forces everything into a restricted and somewhat over-sold perspective of is it like human language or not.”

Central to the whale alphabet study is the use of machine learning techniques to analyze extensive audio data collected by The Dominica Sperm Whale Project. By examining over 8,000 coda sequences from approximately 60 sperm whales in the Caribbean Sea, researchers uncovered previously unseen patterns in whale communication, including codas, which include rubato and ornamentation. 

Similar Patterns In Other Whales

 whale alphabet

Dr. Brenda McCowan, a professor at the University of California Davis School of Veterinary Medicine, underscores the significance of interactive experimentation with whales in unraveling the syntax of their click sequences.

The efforts, combined with observations of whale behavior, offer valuable insights into the communicative abilities of these marine giants.

Machine learning methods may also be helpful when studying vocalizations in other animal species. According to McCowan, elements like tempo, rhythm, and embellishments likely exist in various animal species.

While scientists have observed this complexity in humpback whale songs, indications point to similar patterns among other creatures inhabiting oceans, land, and forests.

Unique Vocal Anatomy

 whale alphabet

Recent studies have shown how baleen whales, including humpbacks, produce their songs underwater. The whales possess a unique vocal anatomy, using a larynx with a specialized structure to create sounds. However, there are challenges posed by human-generated ocean noise, which overlaps with the whales’ vocal frequency range, potentially disrupting their communication. 

New Avenues Of Research

 whale alphabet

While the whale alphabet study’s findings open new avenues for research into animal communication, it is essential to recognize the limitations of machine learning in ascribing meaning to these vocalizations.

As Dr. Rendell emphasizes, while technology can unveil patterns, human interpretation ultimately imbues these signals with significance.

Source: CNN