That title is used in a Washington Post article this week about a recent article in the field of historical linguistics published in the Proceedings of the National Academy of Sciences, which claims to trace a group of words back through seven different language families to a common ancestor language spoken some 15,000 years ago. Historical linguists use cognates (words with similar sounds and meanings) to trace language evolution and are able to compare cognates that appear to be related in a variety of language back to common ancestral languages.  The best-known example is Proto-Indo-European, an ancestor of a number of language families in contemporary Europe and India.

Mark Pagel (University of Reading) and collaborators built a sophisticated statistical model to try to identify specific words across language families that are similar enough, they believe, to have had a common origin. They found 23 “ultraconserved” words. Some which one might expect, including hand, give, I, thou, old and mother, but some which may be surprising such as spit, worm, and bark (of a tree).  The original Washington Post article imagines what might have been said with these words:

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!

It’s an odd little speech. But if you went back 15,000 years and spoke these words to hunter-gatherers in Asia in any one of hundreds of modern languages, there is a chance they would understand at least some of what you were saying. That’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.

As intriguing as this theory is, many linguists are skeptical, as nicely explained by Sally Tompson on the Language Log. So maybe don’t count on being able to communicate with a caveman if you encounter one.