The Divine Spark of Language
A look into the earliest moments of the evolution of language — and what they say about the core of human communication
As the last of the slaves left the great dining hall, and the quiet din of clattering plates and munching mouths took over the room, the court poet — most blessed in the art of language, stood up and began to sing; “I shall recount a time long before heroic Odysseus sailed the ocean blue, a time even before the Greeks were a great peoples on the Earth… It was a dark time before the Fire. We scuttled about the earth like our cousins, the animals. But one day the god Prometheus came barreling down from the sky, chased by watchdogs. He carried with him a giant fennel stalk¹ and inside, a tiny, stolen spark — infant fire from the hearth of Zeus. Since that day we’ve kept predators away during the night, we’ve learned to cook our food, and we’ve built new tools with our smithing. We became humans. We became Greeks.
In a recently published book, Why Only Us², Noam Chomsky and Robert Berwick propose an account of the genesis of language. In a literary flourish, they posit the occurrence of a truly groundbreaking genetic mutation that was so game-changing, and so improbable, that it is almost as if it had been stolen from the divine sphere and gifted to humans. Aptly named, this metaphorical Prometheus passed on these superior genes to their children, who themselves were so much more capable than their cousins that they, as evolution predicts, came to outbreed them. Quickly, this gene spread throughout the species, and Homo Sapiens were changed forever. Like fire, language allowed for the development of completely new social tools and structures that propelled humanity into a new era.
But this version of history is not without critique. Others argue that this story is too mythical, too improbable. To get off the ground, they argue, language doesn’t even need the type of quantum leap that Chomsky seems to think it does. The genesis of language is instead far simpler. I will discuss both these claims in turn, starting with the Chomskian one. But in so doing, we will visit the most fundamental, and possible most unanswerable, questions about langauge. What is it, and where does it come from?
The argument put forth by Chomsky and Berwick in Why Only Us rests on two historical claims, the first cited less confidently than the second. Firstly, there is no evidence that language, as conceptualized by those like Chomsky, (that is, fulfilling the criteria of being generative, having embedded structure and the like,) existed at all before about 80,000 years ago. And secondly, since that initial period of development it hasn’t evolved. Sure, it’s changed — the vocabulary, some linguistic structure (like the order of nouns and adjectives,) the use of verbs to describe motion or path, and much more, differentiate one modern language from another — but no change has fundamentally altered language to the point — using the lingo of evolution — where one could say speciation occurred. Humans across cultures and native languages learn to speak one language or another, yet at the core all these languages share fundamental properties that are exclusive to Homo Sapien communication. We all share them with each other, but not with any other species. Synthesizing these two observations, Chomsky argues that the emergence of language was essentially a one time genetic event — it occurred about 80,000 years ago, gave rise to language as we know it, and hasn’t happened again since.
Turning to evolution, we should first be clear about what it is that has evolved. It is, of course, not languages but rather the capacity for language — that is, UG. Languages change, but they do not evolve. (Why Only Us, Page 91–2.)
UG, Universal Grammar, another theory put forth by Chomsky, is the argument that all human languages share fundamental characteristics that serve as a foundation for all the complexities and intricacies that specific languages, and specific speakers, scaffold on top. This shared foundation allows any human to be born into any language and immediately begin to learn from the noises that constantly surround them. Without this preinstalled software, it would be inconceivable for toddlers to learn so much in such little time with so few examples. And somewhat surprisingly, this superpower is not shared without our primate cousins.
A newborn human infant instantly selects from the environment language-related data, no trivial feat. An ape with approximately the same auditory system hears only noise. The human infant then proceeds on a systematic course of acquisition that is unique to humans, and that demonstrably goes beyond what any general learning mechanism can provide, from word learning to syntactic structure and semantic interpretation. (Why only Us, Page 98.)
In short, Chomsky is making the following claim: Given the empirical observations made about the use of language, it seems that there must have been a fundamental shift away from anything that existed previously. There is only one version — we’re the only species with it, and we’ve only got one type. And in order to get this type of change, you need an explantation that isn’t a continuation of the past, but a clean break from it. A genetic mutation does the trick. It changes enough that it could explain why we’re the only species with the capacity for language, and it’s improbable enough that it wouldn’t have happened twice.
But genetic mutations are tricky things. They’re random occurrences, and as the Law of Entropy would have it, randomness doesn’t usually give way to more complexity. Thus, if this picture is the correct one, the amount of ground broken by the genetic mutation must be minimized — shifting it away from the realm of the impossible into the realm of the improbable. It must be as simple as possible. In a minor slight of hand, Chomsky pivots from the stance of argumentative conclusion, back to premise. He takes his theoretical conclusion (based on, at least what some argue to be, shaky facts,) and uses it in turn to make another claim about the nature of UG:
What evolved, UG, must be quite simple at its core. If so, then the apparent complexity and variety of languages must derive from changes since the shared capacity evolved… (Why Only Us, Page 92.)
It seems to me that however useful Chomsky’s conception of the Prometheus, it might be mired in some circular reasoning. Chomsky et al. have been working for decades to assemble a list of rules that capture Language, not any old spoken language, but the core of it all. This of course is no easy task. Human languages are diverse, divergent and complicated. To come up with any set of generalizable structure requires that its rules be as simple and as basic as possible. These decades of research were guided by a clear mindset:
We may also anticipate, as noted, that the appearance of complexity and diversity in a scientific field quite often simply reflects a lack of deeper understanding, a very familiar phenomenon. (Why Only Us, Page 93.)
This structural simplicity allows for Chomsky to make the bold claim that he does — it’s simple enough that it could have come about by chance. But to then turn around and say that a simple genetic event is evidence that the UG must therefore also be simple — seems a bit circular to me.
There are, of course, those out there who wholly disagree with Chomsky. Not just because the idea of a genetic mutation happening all at once, and giving rise to something as complex as language. But more so because they don’t believe that such a drastic explanation is necessary. If it were the case that there was something wholly unique about human language, Chomsky and his theories might have a fighting chance. But there isn’t anything too unique. Regular old, gradual, evolutionary changes work just fine — language speciation never occurred.
In a recent TEDx Talk entitled The Origins and Evolution of Language³, psychologist Michael Corballis argues against the Chomskian hypothesis I’ve run through above. At it’s most basic, he thinks that the impossibility of such a mutation makes the Chomsky account biblical-esque, insisting that a better name for that individual would might be found there:
He doesn’t call that individual Adam, he calls that individual Prometheus. For some reason, known to him… That’s a biblical story really, because it says that langauge happened in a single individual, once, with a single mutation of some sort, and I don’t think it makes biological sense. (3:24–50 into the clip.)
He seems to have clearly missed the point about Prometheus, which hopefully I’ve made clear enough above. Nonetheless, his point stands. The Chomskian picture is a bit far fetched. And if true, might take evolutionary theory as a casualty. If such a complex ability were to arise all at once, then what’s to say that other complexities of biology couldn’t similarly arise all at once, and not over eons, as Darwin supposed?
Corballis posits another view, one that requires less of a leap. Language is not something new, but something evolved from capacities that our primate cousins have as well.
There is an area in the monkey brain that is activated whenever the monkey reaches out to grasp something. But not only then. It’s also activated when it sees other monkeys, or even humans, doing the same action, for which it was given the name The Mirror System. And it seems that the areas of the human brain responsible for language are in fact the same parts of the brain as the Mirror System — the Mirror system has developed into the language system. And with enough observation of great apes in the wild it seems that they use gesturing as a means of communication, one that it far more expressive then there vocal calls, and one that rests on the verge of being language-like. Further, though no primate has ever been taught to speak, individuals have been taught to use sign language. When interacting with humans, these individuals can seemingly be pushed to learn language — it’s nothing too fancy or descriptive, but it seems just enough to wonder if the primate brain is just capable enough for language. After all, a language consisting of signing, in some circumstances, can be a fully-featured language. Many deaf people around the world make use of quite complex and modern sign langauges.
But of course, there is something unique about human language. It is far more expressive. But more importantly, it’s vocal. Corballis argues that it was bipedalism that was fuel to get human language off the ground. As humans became increasingly more bipedal (starting millions of years ago, rather than Chomsky’s birth date of language at tens of thousands of years ago,) newfound freedom for our hands and visibility for our faces allowed for gesturing languages to become more complex. Over time these gestures became increasingly more complex, and then after a while, more simple and standardized. Vocalizations slowly became more prominent, as they allowed for more expressive communication that wasn’t limited by line of sight, and also once again freed the hands for other tasks.
Corballis makes a significant but underplayed aside in the middle of his talk:
I should add, by the way, that the idea that there was a sudden event is widely held not just by Chomsky, but by a lot of archeologists. Although most biologists I talk to think that it doesn’t make sense. (12:13 into the clip.)
It seems that there’s evidence that something drastically changed about 80,000 years ago. Something that can be picked out by the archeological record and something that indicates a fundamentally change to the way humans use language. Evidence that Chomsky surely sees as supporting him. Corballis makes little of it, but it seems to me that if he were to reduce it’s efficacy at supporting Chomsky, he might argue that it was at this time that human language became truly modern. It wasn’t a break from the past, but simply one moment along a progression that, in retrospect, left a mark.
But a plausible explantation for language’s gradual development is not sufficient to refute Chomsky’s claims. He argues that there is something unique to human language not shared with any other species, even with our primate cousins. Namely, human language makes uses of embedding, which allows for infinite productivity, while other species, even if they engage in language like behavior, employ only finite state grammars or the like. Thus, according to Chomsky, by making a claim like Corballis’, one must also give up on the simply described difference betwen human and non-human langauge. Human language’s complexity in relation to animal language must be a great difference in value, rather a small difference in kind, something which Chomsky finds implausbible. Corballis might find this alternative quite agreeable — either arguing that human language is in fact nothing too special, that our primate cousins can construct linguistic embeddings themselves, or that though embedding is a cute description of human language, it is only a theoretical one. Instead we, like our ape cousins, don’t possess all too fancy tools for language.This is to say that one’s position about the historical reality of the development of human language also interplays with the possible range of capacities of human language — and animal language. They go hand in hand.
There have been several studies on different species ability to understand recursion, from primates, to songbirds, to humans. It seems less clear than Chomsky would have it that humans clearly possess recursion, and that other animals do not. There is evidence that points against both of those claims.
In sum, it seems that there is much left to be discovered about the nature and history of human language. When and how it came about are still very much live questions, not only because we don’t have a clear picture of the past, but because we aren’t even sure what language is.
To conclude this blog post, I’ve attached two videos of animals (note in the second one, the animal isn’t even a primate,) that carry out complex ‘recipes’ to achieve a certain goal. Though not a linguistic task, these plans do seem to be embedded. ‘I don’t have the peanut.’ ‘I eat the peanut.’ Embedded between the two is ‘I get water, and pour it into the tube.’ If these are cases of recursion, then it seems that other animals are clearly capable of a task Chomsky would argue only humans possess.
Citations
- Hesiod. The Homeric Hymns and Homerica with an English Translation by Hugh G. Evelyn-White. Theogony, Line 565. Cambridge, MA.,Harvard University Press; London, William Heinemann Ltd. 1914.
- “Chapter 3: Language and Evolution.” Why Only Us: Language and Evolution, by Robert C. Berwick and Noam Chomsky, The MIT Press, 2016.
- TEDx. (2018, August 14). Michael Corballis: The Origins and Evolution of Language [Video file]. Retrieved from https://www.youtube.com/watch?v=PLXsUgL_wpc
