Animal Speech Isn't As Random as Once Thought: The Intricacies of Language

It turns out that animals may have more complicated "speech" patterns than once thought. Scientists have found that the calls of many animals, from whales to wolves, may contain more language-like structure than previously thought.

Many species of animals produce complex vocalizations. For example, the mockingbird can mimic over 100 distinct song types of different species. But while these vocalizations seem to suggest language-like characteristics, scientists have had trouble decoding the complexity of these calls.

In fact, scientists have long assumed that the sequence of animal calls is generated by a simple, random process, called a "Markov process." This means that the vocal elements are dependent only on a finite number of preceding vocal elements, making the process fairly random and far different from human language.

Human language uses what is called "context-free grammars." This is when certain grammatical rules apply regardless of context. Animal language, though, uses simple or "regular" grammar, which is far more restrictive. The Markov process is the most common model used the examine animal vocal sequences, which assumes that a future occurrence of a vocal element is entirely dependent by a finite number of past vocal occurrence.

Yet this may not be the case. The researchers found no evidence for a Markovian process after using mathematical models to analyze the vocal sequences of chickadees, finches, bats, orangutans, killer whales, pilot whales and hydraxes. In fact, most of the vocal sequences were more consistent with models that are more complex than Markov processes.

"Language is the biggest difference that separates humans from animals evolutionarily, but multiple studies are finding more and more stepping stones that seem to bridge this gap," said Arik Karshenbaum, lead author of the new study, in a news release.

The findings are published in the journal Proceedings of the Royal Society B: Biological Sciences.