Shifty Talk: Probing the process of word evolution

Here’s an evolutionary talking point: Two new studies quantify parts of the mechanism by which frequently used words change slowly over many millennia whereas rarely used words more rapidly take on new forms.

In fact, frequency of word usage exerts a “lawlike” influence on the rapidity of language evolution, the research teams conclude in the Oct. 11 Nature. This discovery offers a new tool for retracing the history of major language families, reconstructing ancient tongues, and predicting which words will undergo future alterations.

“We expect all languages to diverge initially in the least frequently used parts of their vocabulary,” says evolutionary biologist Mark Pagel of the University of Reading in England.

Pagel’s group focused on Indo-European languages. Some words for the same meanings differ strikingly across the more than 100 languages and dialects of that family, while others take similar forms.

The researchers first determined 200 basic vocabulary meanings in 87 Indo-European languages spoken during the past 6,000 to 10,000 years. They then applied a statistical technique to modern-language data in order to estimate the spoken frequencies of the corresponding words in English, Spanish, Russian, and Greek.

Among those 200 meanings, commonly used words—such as who or night, and terms for numbers—evolved slowly and sounded similar in different languages. Such words undergo no more than one wholesale shift to a new form every 10,000 years, the scientists propose.

In contrast, less frequently used words—such as dirty, turn, and guts—evolved more rapidly and sounded different across languages. These types of words change forms up to nine times every 10,000 years, according to the investigators.

In the second new study, Harvard University genomics graduate student Erez Lieberman and his coworkers measured the rate at which English verbs have become regular—using the suffix “ed” to signify past tense—over the past 1,200 years. That linguistic period begins with Old English, includes Middle English around 800 years ago, and ends with English as it is spoken today.

The team compiled a list of 177 irregular verbs in Old English. Of that number, 145 remained irregular in Middle English and 98 are still irregular today.

The researchers then calculated the frequency of each verb’s usage in Modern English and estimated frequencies for the two older tongues. They determined that an irregular verb used 100 times as often as another in daily conversation takes 10 times as long to become regular as the less-spoken verb does.

If current trends continue, only 83 of the 177 verbs studied will be irregular in 500 years, the researchers predict. They predict that the next irregular verb to regularize will be wed, meaning that just-married couples will no longer be “newly wed” but will have blissfully “wedded.”

“Our results indicate that languages can evolve in such an orderly fashion that simple mathematical descriptions capture their behavior,” Lieberman says. “A language’s irregularities reveal the mechanisms shaping its evolution.”

The use of sophisticated statistical methods to quantify how words evolve on the basis of the frequency of their use “is an important step forward,” remarks psycholinguist W. Tecumseh Fitch of the University of St. Andrews in Scotland.