A recent study of an ancient language provides new insights into the nature of linguistic evolution, with potential applications for today's world. The study, "Dvandvas, Blocking, and the Associative: The Bumpy Ride from Phrase to Word," to be published in the June 2010 issue of the scholarly journal Language, is authored by Paul Kiparsky of Stanford University.
Dr. Kiparsky's research focuses on the reasons why languages change over time, and the mechanisms by which this change occurs. Linguistic change differs from biological evolution and socio-cultural change because of the way language is organized and learned. Languages are passed on by example, but each is governed by a coherent set of rules that conform to a common set of organizing principles.
Linguistic change is typically initiated by children as they make "intelligent" errors in seeking the simplest way of navigating the languages they are learning. By studying linguistic change, we gain new insights into how language is organized and how children learn language.
Dr. Kiparsky observed that linguistic change does not follow a straightforward path toward a simpler system. Instead, it takes a "bumpy ride" to its destination. A language is like an enormous house that has to be reconstructed by each new occupant, who has to discover its design as the work is in progress, and while the previous occupants are still living in it.
Construction is always going on, now and then a room is finished, but only after centuries can an outside observer see that a fundamental renovation has taken place. Dr. Kiparsky's new study shows how the stepwise progress of innovations through a language follows an orderly course predicted by principles that appear to be shared by all languages.
This new insight into the nature of how language change occurs will help linguists and those who rely on their research to gain a greater understanding of language and the mind.
Explore further: Seven projects to make progress on ethics and global food security in five years
More information: A preprint version is available on line at: