Do sperm whales have a culture?

For decades, the idea that non-human animals might have a “culture” was treated as a romantic projection with no scientific basis. Culture, it was said, was by definition human: a body of knowledge, behaviours and values shared by a group, passed on by social learning, and varying from one group to another independently of genetics. That definition assumed tools, a language, explicit teaching. Not marine animals. Then, from the 2000s onwards, long-term observations off the Galápagos, the Azores, Dominica and Sri Lanka began to draw a different picture. The sperm whale, Physeter macrocephalus, is today regarded by most specialists as a textbook case of animal culture.

The social architecture, in four levels

To understand sperm whale culture, one must first grasp its social organisation, which unfolds over four nested levels. At the smallest level is the social unit: a stable group of six to twelve related females and their young, which stays together for decades, sometimes for life. It is the equivalent of the family clan, and daily cooperation, allonursing, the guarding of calves at the surface while the mothers dive deep, all of this happens at this scale.

At the level above, the group temporarily brings together several social units. These gatherings last from a few hours to a few days, for the duration of a productive feeding area or a breeding opportunity. The units then separate and resume their own course. The links between units are at once historical (shared genetic relatedness), geographical (zones of overlap) and linguistic.

At the third level, the clan, the cultural dimension begins. A clan is a set of social units sharing the same coda dialect. These units may never meet physically; a clan extends over thousands of kilometres. In the eastern Pacific, six main clans have been identified and named after the rhythmic characteristics of their codas (Short, Plus-One, Four-Plus, Regular, Slow, Rapid). A recent estimate (Whitehead & Rendell, 2024) suggests that a North Pacific clan could number up to twenty thousand individuals spread over an entire ocean basin.

Finally, at the fourth level, the vocal lineage encompasses clans that share certain acoustic traits. It is the acoustic equivalent of a language family, in the way that Indo-European encompasses French and Hindi.

Codas as a dialect

The term “coda” refers to a rhythmic sequence of three to twelve clicks produced by a sperm whale at the surface or shortly after. Unlike echolocation clicks, which are intense and focused, codas are relatively quiet and are exchanged at short range between the members of a unit. Each clan has its own coda repertoire: characteristic rhythmic patterns that distinguish it from the other clans.

The rhythmic signature is strikingly precise. The “Regular” clan of the eastern Pacific uses codas whose inter-click intervals are evenly spaced. The “Plus-One” clan systematically ends its codas with an offset click. The “Short” clan uses briefer sequences. These differences are not incidental: they are stable over decades, passed on by learning, and function as a genuine marker of identity. When two units belonging to different clans cross paths, they avoid each other socially and do not form a mixed group.

More than sounds: shared behaviours

The boundary between clans is not only acoustic. Clans also differ in:

• Hunting techniques: preferred prey, typical dive depths, the duration of breath-hold cycles;
• Migration routes: the units of a single clan tend to use the same ocean corridors from one year to the next;
• Differential breeding success according to climatic years. Whitehead & Rendell have documented that in El Niño years some clans collapse reproductively while others thrive, although they swim in the same waters. Culture therefore partly determines survival.

How the culture is passed on

The mechanism of transmission is essentially matrilineal and long-term. A young sperm whale stays with its maternal unit for between ten and fifteen years. During that period it witnesses thousands of coda exchanges, watches hunting techniques and accompanies the unit on its travels. Its brain, one of the largest in the animal kingdom (7.8 kg on average, the heaviest known), is massively oriented towards acoustic and social processing.

At puberty the males leave; they join “bachelor groups” of young adults, then become solitary as they migrate towards high latitudes. They will return sporadically to tropical waters to breed. The females, for their part, remain for life in the unit they were born into. This structure stabilises the vertical transmission of cultural capital over decades, even centuries. A coda dialect observed in a clan today may be identical to the one recorded forty years earlier in the same area.

Towards a language? The CETI project

All of this raises the obvious question: are these codas a language? The question is a delicate one. Human language has properties (combinatoriality, recursion, symbolic reference, syntax) that are difficult to demonstrate in other species. But recent results blur the boundary. The Cetacean Translation Initiative (CETI), launched in 2020 off Dominica, applies artificial intelligence methods to several hundred hours of coda recordings. A study published in 2024 by Pratyusha Sharma and her team at MIT (Nature Communications) brought to light a combinatorial structure in codas: systematic variations of rhythm and tempo that resemble modulable “phonemes”. The vocabulary identified would be far richer than had been thought.

Another study (Hersh et al., 2022, PNAS) had already suggested that certain codas may have a function of symbolic marking: a kind of clan identity flag. If these results are confirmed, sperm whales would have at their disposal a structured communication system, culturally transmitted, and carrying collective identity. Everything that distinguishes a language from a repertoire of instinctive calls.

Sources

• Whitehead, H. & Rendell, L. (2015). The Cultural Lives of Whales and Dolphins. University of Chicago Press.
• Rendell, L. & Whitehead, H. (2003). “Vocal clans in sperm whales (Physeter macrocephalus)”. Proceedings of the Royal Society B, 270.
• Sharma, P. et al. (2024). “Contextual and combinatorial structure in sperm whale vocalisations”. Nature Communications, 15.
• Hersh, T.A. et al. (2022). “Evidence from sperm whale clans of symbolic marking in non-human cultures”. PNAS, 119.
• Whitehead, H. (2003). Sperm Whales: Social Evolution in the Ocean. University of Chicago Press.

---

Further reading