IACSIACSInt'l Academy for Consciousness Studies
Do animals have minds?

Do animals have language?

The short answerAnimals communicate richly, but most scientists say they do not have language in the full human sense. Vervet monkeys give predator-specific alarm calls, and trained apes and parrots learn hundreds of symbols. What no animal reliably shows is grammar: combining a small set of words into endless new, hierarchically structured sentences. That recursive syntax appears uniquely human.

How rich animal communication really is

Animals communicate constantly, and some of it is remarkably sophisticated. Vervet monkeys, studied by Dorothy Cheney and Robert Seyfarth, give different alarm calls for leopards, eagles, and snakes, and other monkeys respond appropriately to each: running up a tree for a leopard call, scanning the sky for an eagle call. Honeybees encode the direction and distance of food in a waggle dance. Prairie dogs, songbirds, and dolphins all carry meaningful information in their signals.

Trained animals go further still. The bonobo Kanzi learned to use hundreds of visual symbols called lexigrams and to understand many spoken English requests. Alex, an African grey parrot trained by Irene Pepperberg, could label colors, shapes, and numbers and answer simple questions. The chimpanzee Washoe learned signs adapted from American Sign Language. These achievements are real and show that animals grasp symbols, meaning, and reference. The debate is not whether animals communicate, but whether what they do counts as language.

The one thing that seems uniquely human

Most linguists draw the line at grammar, specifically recursion: the ability to combine a limited set of words into an unlimited number of new sentences, and to nest phrases inside phrases (the dog that chased the cat that caught the mouse). In an influential 2002 paper, Marc Hauser, Noam Chomsky, and Tecumseh Fitch argued that this recursive syntax may be the core feature that sets human language apart from all animal communication systems.

Animal signals, by contrast, tend to be holistic: each call stands for one thing and calls are rarely combined into structured new meanings. Vervets never join their leopard and eagle calls to say something new. Even Kanzi, the most linguistically gifted ape on record, produced mostly short combinations without the flexible, rule-governed structure a human toddler acquires effortlessly. Reviews of the evidence, including a 2021 survey of language-trained animals, conclude that no non-human has demonstrated open-ended hierarchical syntax. Notably, several early claims of animal grammar, including some vervet findings, were later walked back by the researchers themselves.

So what is the honest answer?

It depends on what you mean by language. If language means communicating meaningfully, then many animals plainly have it. If it means the specific human system of unlimited, grammatically structured expression built from recombinable units, then the evidence says no animal has been shown to have it. Human language is best understood as one branch on a rich tree of animal communication, distinguished less by having meaning than by its combinatorial power.

This remains an active field, not a settled verdict. Researchers continue to probe whether some species combine signals in simple grammar-like ways, and new studies of whale and bird vocalizations keep testing the boundary. But the current mainstream view is clear: animals are eloquent communicators, and human language is a special case whose defining trick, endless structured recombination, has not been found elsewhere.

Sources and further reading

Go deeper at IACS
Study this question with an AI professor who has spent a lifetime on it.
All of the Big Questions