Skip to content
NatureHQ

Claim check

Do whales have dialects?

SupportedGood evidence backs this, and specialists broadly agree.

Yes, in the sense biologists use the word: stable, group-specific vocal repertoires learned from other members of the group. Killer whale pods keep distinct call repertoires for decades, and sperm whale clans sharing the same water go on sounding different. It says nothing about whether the calls mean anything.

The claim as it circulates

“Whales from different regions speak in different dialects, like human accents.”

Where you may have met it: Documentary narration about killer whales and sperm whales; Google’s own search suggestions around whale and orca communication; Popular writing about animal culture, usually without defining the word

Taking it apart

What was claimed
That whale populations have dialects comparable to regional human accents.
What was actually observed
Resident killer whale pods were recorded over more than a decade and found to use stable repertoires of discrete calls, with repertoire similarity tracking recent maternal ancestry rather than geography. Separately, sperm whale coda repertoires sort groups into a small number of clans whose ranges overlap without their repertoires converging.
What the evidence supports
That group-specific vocal repertoires exist, persist across decades, and are transmitted socially. The sperm whale case is the stronger evidence for learning specifically, because clans sharing the same water and staying distinct rules out both geography and habitat as explanations.
What it does not support
That the calls carry meaning, or that a dialect difference is like a language difference. The function of individual call types is largely unknown in both species. In killer whales, repertoire is shared along matrilines, so learning and inheritance are correlated and hard to separate completely.

This is a case where the technical term and the everyday one overlap enough that the claim survives, provided the borrowed word is unpacked. A dialect in biology is a shared, learned, group-specific repertoire. It carries no implication that the calls mean anything, and human accents differ in the pronunciation of words that do.

The killer whale evidence is a decade of recording individually known pods. Each pod uses its own set of discrete calls, the sets stay distinguishable over many years, and pods that share recent maternal ancestry share more calls than pods that merely share water. Stability that long rules out drift; the pattern of sharing points at learning down the maternal line.

The sperm whale evidence is stronger in one specific way, and the two cases cover each other’s weakness. Because orca pods are matrilines, genetics and learning travel together. Sperm whale clans do not have that problem: clans with quite different coda repertoires use the same patch of ocean, and go on sounding different. There is no distance between them and no difference in the water, which leaves social learning holding the result.

The claims underneath

Each one carries its own evidence, scope and caveats. Expand any of them to reach the studies.

Each resident orca pod uses its own set of calls, passed down the maternal line and stable for decades.

Established

Specialists would state this without hedging. Multiple independent lines of evidence agree.

Resident killer whale pods maintain stable, group-specific repertoires of discrete call types. Repertoire similarity between pods tracks recent maternal ancestry rather than geographic proximity, and individual repertoires persist over decades.

Who this applies to
Established in resident (fish-eating) killer whales of the north-eastern Pacific. Other ecotypes and regions are less well characterised.Do not extend this beyond the taxa listed — the popular version over-reaches.
Studied in
Orcinus orca
Why we rate it this way, and what the caveats are
EstablishedHigh confidence

Continuous recording of individually known pods over more than a decade, with repertoires that remained distinguishable throughout, and a pattern of sharing that follows matrilines rather than geography.

How far it can be extended

Resident, transient and offshore ecotypes differ substantially in social structure and acoustic behaviour.

Caveats

  • Because pods are matrilines, learning and inheritance are correlated and hard to separate completely.
  • What individual call types are for is largely unknown; the evidence is about repertoire, not meaning.

Still unanswered

  • The function of specific call types.
  • How much repertoire a calf acquires from its mother versus from the wider pod.

Last reviewed 2026-08-31

The evidence (2 studies)

Sperm whales sort into clans that sound different from each other, and clans sharing the same water keep sounding different.

Well supported

Good evidence backs this, though some details remain open.

Sperm whale social groups cluster into vocal clans distinguished by coda repertoire. Clan boundaries do not track geography, and clans with different repertoires occupy overlapping ranges without repertoire convergence, which is inconsistent with habitat-driven or purely genetic explanations.

Who this applies to
Established most firmly in the South Pacific and the Eastern Caribbean, where long-term recording programmes exist.
Studied in
Physeter macrocephalus
Why we rate it this way, and what the caveats are
Well supportedHigh confidence

The overlap result is what carries it: two clans in the same water staying acoustically distinct removes the ordinary environmental and geographic explanations, and has held up across basins and decades.

How far it can be extended

Clan structure has been described in several ocean basins, but the detailed evidence comes from a small number of studied regions.

Caveats

  • Clan membership is inferred from repertoire similarity, and the boundary depends on how the clustering is done.
  • Codas are recorded from groups more often than from identified individuals.

Still unanswered

  • How a calf acquires its clan repertoire, and how early it is fixed.
  • Whether clan membership has consequences beyond acoustics, such as in foraging success.

Last reviewed 2026-08-31

The evidence (2 studies)

Go deeper

Orca

Females stop breeding at around forty and can live to eighty — and lead the hunt when food runs short.

Related things people ask

Last reviewed 2026-08-31