Science-- there's something for everyone

Thursday, April 25, 2013

The development of human language

Language is the hallmark characteristic that sets humans apart from other animals. More than tool use, empathy or morality, all of which are practiced by at least some non-humans, language makes us who we are. At some point, we evolved the ability to turn a few dozen sounds into a limitless number of expressions. Charles Yang of the University of Pennsylvania tried to figure out when that happened by comparing two linguistically similar creatures: very young children and chimpanzees. You won’t be surprised to learn that they aren’t that similar after all.

The big question in language acquisition is how young children get from speaking no words to complex sentences so quickly and accurately. There are two prevailing ideas. One is that toddlers begin their journey into language use with imitation. That is, they simply repeat the short phrases that they hear adults say. Only after mastering those sentences do they go on to improvise their own longer sentences. The second idea is that children combine language elements independently from the very beginning, based on the grammar of the speakers around them.

To evaluate these two possibilities, Yang noted how often young language learners, speaking only two-word sentences, used ‘a’ or ‘the’ before nouns. He compared this ratio to that found in the Brown Corpus (a collection of English language texts) and with over a million utterances appearing in the public domain that were directed at children. In adult speech, certain words tend to be paired almost exclusively with one or the other of these determiners (we almost always refer to ‘the kitchen’ rather than ‘a kitchen’), but young children used the two articles much more equally. This strongly suggests that even at the very earliest stages of language acquisition, they are not simply parroting back phrases they’ve previously heard.

Obviously, chimpanzees don’t speak, but a few of them can sign. Do they also combine signs independently of ones they’ve seen? Here, Yang uses a sample size of one: the American Sign Language-using chimp named Nim Chimpsky (after noted linguist Noam Chomsky). Unlike the human kids, Nim’s language skills seemed to be based purely on memorization. He could not improvise new combinations of signs.

This corroborates that there really is something unique about human language. At some point after we diverged from chimpanzees, we evolved the ability to communicate in a fundamentally different way. How that happened is still a mystery, and may remain so, since we can look to neither the fossil record nor our living cousins for answers.

Yang, C. (2013). Ontogeny and phylogeny of language Proceedings of the National Academy of Sciences DOI: 10.1073/pnas.1216803110.


  1. I remember reading about a reseacher who, wanting to study something along similar lines, paired his son with a chimpanzee but soon had to abandon the experiment as instead of the chimp being carried along by his son, his son started to regress and act more like the chimp. Which I found mildly amusing. :)

    Also, have you ever read any of Chomsky's language books? I got a headache just reading the introduction of New Horizons in The Study of Language and Mind! I think Spike Milligan is just about my level :)

  2. I haven't read Chomsky, but I have read Steven Pinker's book "The Language Instinct". It makes the same case as the authors of this study, namely that children don't learn language simply by imitation.