One of the most interesting stories of last week came from research by linguists suggesting that some words have descended largely unchanged since the end of the last Ice Age. This research – from Mark Pagel, an evolutionary theorist at the University of Reading in the UK, along with Quentin D. Atkinson, Andreea S. Calude, and Andrew Meadea – describes a method for discovering ancient language superfamilies. The paper appeared on May 6, 2013 in the Proceedings of the National Academy of Sciences. These researchers’ method for discovering language superfamilies uses cognates – that is, two words with a common origin – and statistical analysis of existing languages. The research led the Washington Post to describe a sentence that might be translated into any one of hundreds of modern languages and still be at least partially understood by hunter-gatherers in Asia some 15,000 years ago. That is:
You, hear me! Give this fire to that old man. Pull the black worm off the bark and give it to the mother. And no spitting in the ashes!
How is it possible it would be understood in various languages, across time? As the author of the Post’s article, David Brown, said, it’s because:
… all of the nouns, verbs, adjectives and adverbs in the four sentences are words that have descended largely unchanged from a language that died out as the glaciers retreated at the end of the last Ice Age. Those few words mean the same thing, and sound almost the same, as they did then.
Linguists group languages into families based on similarities of grammar and vocabulary. Cognates like father (English), patêr (Greek) and pitr (Sanskrit) indicate that a now-lost language divided thousands of years ago into separate branches including English, Greek, and Sanskrit (plus many more). This is the Indo-European family, probably the best known. But there are hundreds more language families.
Can the roots of language families be traced back even further, to superfamilies? The work of Pagel and colleagues suggests it can, but that answer is controversial. After all, how is it even possible to know? Language experts tell us that most words disappear and are replaced about every 2,000 to 4,000 years. More heavily used words can last (be conserved) up to an estimated 9,000 years.
But can any word be ultraconserved – that is, last as long as 15,000 years? Pagel and his team say yes. And, they say, finding those words and their present-day cognates across a variety of languages could reveal very large-scale language superfamilies.
To find these key words, they first carried out a painstaking statistical analysis of six language families. Based on frequency of use, they zeroed in on words likely to show what they called deep ancestry.
Next, they applied the method to major language families: Altaic, Chukchi-Kamchatkan, Dravidian, Inuit-Yupik (Eskimo), Indo-European, Kartvelian, and Uralic. And they found their ultraconserved words – what they called “immortal” words – and cognates, indicating that these families might all descend from a long-gone superfamily, which Pagel calls the Eurasiatic superfamily.
So what are the words? There are nearly 200. The most frequent found in multiple languages, listed by number of cognates, are:
7 – thou (obsolete singular form of you in English)
6 – I
5 – not, that, we, to give, who
4 – this, what, man/male, ye (obsolete plural form of you in English), old, mother, to hear, hand, fire, to pull, black, to flow, bark, ashes, to spit, worm
The paper does not list the likely pronunciation of the protowords, that is, the earliest, first or original words. However, you can hear a comparison of a few of the words in modern languages at this fascinating page from the Washington Post.
The team arrived at the estimated date of 15,000 years ago by dating the protolanguages, for instance proto-Indo-European, by determining the differences between modern descendants. The same technique uses the difference between the protolanguages themselves.
The superfamily theory tends to stir up strong feelings in academia. William Croft, a linguist at the University of New Mexico – Albuquerque, was quoted in sciencemag.com as saying:
It probably won’t convince most historical linguists to accept the Eurasiatic hypothesis. But their resistance may soften somewhat.
Bottom line: Research from linguist Mark Pagel and colleagues describes a method for discovering ancient language superfamilies and suggests that some words have descended largely unchanged since the end of the last Ice Age.