Saturday, February 26, 2011

Chatting with Cavemen

The ancients understood very well that languages were related to one another in groups such as Celtic and Germanic. Whether they recognized broader interrelationships and common origins is unknown. The rigorous study of these relationships is usually dated to lectures on the subject by Sir William Jones in 1786. In the two centuries since then, we’ve come to understand the self-consistent ways languages diverge and evolve, so often it is possible to discover a common origin for languages that might not have many obvious similarities. In the case of Indo-European languages (which include English, Greek, Russian, and Hindi, among many others), researchers actually have reconstructed an ancestral proto-Indo-European language more than 5000 years old with a vocabulary of hundreds of words. African, East Asian, and Native American languages also have been categorized into a handful each of large language groups.

A controversial question is whether these broad language groups in turn also have a common origin. Was there a single mother tongue? There are tantalizing clues. Researchers at the University of Reading have used supercomputers to cross-analyze languages and believe they can trace some words back 30,000 years. Even a cursory comparison of vocabularies in widely dispersed languages turns up spooky similarities. Take the word “who” as single example among many: Khoisan, “!khu”; Latin, “qui”; Austric, “o-ko-e”; Dene-Caucasian, “kwi”; Amerindian, “kune”; and so on.

Linguist Merritt Ruhlen finds the evidence for a single mother language more than compelling. I recommend his book The Origin of Language: Tracing the Evolution of the Mother Tongue as a readable and cogent argument by a radical unifier. The unifiers have been making substantial academic headway lately, but still are outnumbered by conservatives. Conservative linguists don’t so much deny a monolingual origin as say it cannot be reconstructed if it existed and that coincidence is an explanation for similarities among unrelated tongues. Unifiers don’t buy “coincidence” for what they see as far too many correspondences among supposedly unrelated languages. Some roots Ruhlen proposes as coming from an original language:

ku= “who”
ma= “what”
pal= “two”
akwa= “water”
tik= “finger” (as in English “digit”)
bungku= “knee”
putv= “vulva”
chuna= “nose, smell”

We don’t know when humans first spoke a full-blown language, defined as words with discrete meanings strung together with a true grammar. (How chatty the Neanderthals may have been is a question that starts angry arguments among paleontologists, but nobody really knows.) It is all but certain, though, that by the time anatomically modern humans spread out from Africa 100,000 years ago or so, they happily were bragging and gossiping. We should keep in mind just how small the modern human population was at this time: a few tens of thousands at most. This is one of the best arguments for the monogenesis theory of language: a few thousand people populated the entire world. We may not have much in common with the denizens of that era, but it’s nice to think we could have exchanged a few words.

Post Script

Two tales of my own featuring talkative ancestors: Neander Valley Girl (or Cavemen Behaving Badly) at , set 35,000 years ago, and Modern Times at , set about 100,000 years earlier and 9000 kilometers to the south.

1 comment:

  1. This is something I've wondered about, but didn't know they were actually trying to reconstruct. Think I'll check out that book.