Listen to a narrated version of this essay:
Human language combines sounds to form words, which combine to create meaningful utterances. But not all combinations are equally acceptable. Brqqwmxd, for instance, both looks and sounds (or at least looks like it would sound) like a terrible English word: were we to coin a new word for, say, the lightness we feel in our legs after taking off skis, brqqwmxd is the last word we’d choose. Or, to put it less grammatically, brqqwmxd is we’d last the choose word.
But do nonhuman animals have grammar? Grammar dictates how linguistic elements can be combined. Grammatical rules bring structure—hierarchy—to a language. And no doubt about it, hierarchical structure can be found in the songs and calls of plenty of other species. Take the long mating songs of male humpback whales, which have been shown to be composed of repetitive “phrases” embedded inside longer, repetitive “themes.”
Or consider Bengalese finches, which in a recent study were conditioned to ignore the individual songs of other finches after listening to them hundreds of times. When the birds were then played versions of those songs in which elements had been scrambled by the experimenters, one—but not the rest—of the scrambled songs elicited strong responses from nearly all of the finches. The researchers interpreted this as evidence that this version had violated the rules of birdsong, which in turn suggests that birdsong has rules and that finches are capable of learning them. (The researchers cleverly supported their interpretation in a subsequent study that exposed long-suffering finches to an artificial language composed of syllable strings; these finches, too, were later able to distinguish between novel strings that followed the rules of the artificial language and novel strings that did not.)
But what is missing—what human language has and the communicative attempts of other species, no matter how valiant, do not—is combinatorial rules at the level of meaning. That is, in human language, it’s not just sounds that combine systematically; meanings themselves combine to create more complex meanings. Finch combines with long-suffering and the suffix –s to produce long-suffering finches, a phrase whose meaning is not only derived from—but, thanks to syntax, surpasses—the sum of the meanings of its parts. No incontrovertible evidence exists, however, that when two nonhuman animal calls are conjoined, the meaning of the new sequence similarly builds on the meanings of its components.
One possible exception comes from research on wild male Campbell’s monkeys in the Ivory Coast. According to researchers, these monkeys have six distinct calls. Four of the six are short, one-element calls, two of which serve as alarm calls for specific predators: hok for crowned eagles, and krak for leopards. But two additional calls are created when monkeys append a single, invariant sound to the end of each—hok-oo and krak-oo. Unlike hok and krak calls, which are given only in very specific situations, hok-oo and krak-oo are given under a wide range of circumstances. (Specifically, the researchers report that hok-oo calls “are given to a range of disturbances within the canopy, including eagles, the presence of neighboring groups and, on a few occasions, to a flying squirrel,” while krak-oo calls are even more generic and “can be given to almost any disturbance”.)
The researchers argue that this invariant addition –oo is “functionally equivalent to suffixation in human language.” In other words, just as adding an –s to the end of a noun changes its meaning in a predictable way (by pluralizing it), adding –oo to a species-specific alarm calls also changes its meaning in a predictable way—perhaps by generalizing it.
Suffixation? Holy finches, that’s grammar! But alas not everyone is convinced. Linguist James Hurford argues in his book The Origin of Grammar, for instance, that –oo may not serve a grammatical function at all. Instead, hok-oo and krak-oo may simply be basic units of meaning in their own right. With such small monkey vocabularies—and the calls in question used in such a wide range of contexts—it will probably prove impossible to tell the difference. In the future, we’ll have to turn to other members of the animal kingdom to determine whether reports of grammatical animal calls have been greatly exaggerated.
