skip to Main Content

UH Manoa Linguists Create New Tool for Tracking Language History

University of Hawaiʻi
Joseph E. Grimes, (808) 695-8402
Linguistics Department
Arlene Abiang, (808) 956-5637
External Affairs & University Relations
Posted: Jan 14, 2005

HONOLULU — The University of Hawaiʻi at Mānoa Linguistics Department, including adjunct professor Joseph Grimes and graduate students Maria Faehndrich and Christine Hansen, in partnership with local high tech firm DataHouse, Inc., have created a new tool for organizing massive data needed to analyze the history of language families.

The tool, called "Wordcorr," balances two principal problems comparative linguists face. Linguists easily recognize patterns of similarity in groups of languages, but have difficulty in keeping track of all the evidence for those patterns. Computers, which are great at keeping track of the details, are unreliable when it comes to recognizing the relevant patterns.

"With ʻWordcorr‘, we can manage more complex language data more accurately than we ever could by hand," said Grimes, who serves as principal investigator on the project. "Besides its use as a heavy duty research tool, it is beginning to be used as a medium for teaching historical linguistics."

The new technology allows linguists to enter data on any group of languages in the world using the International Phoenetic Alphabet, a uniform and universally understood system for transcribing the speech sounds of all languages. Linguists then annotate the data to highlight the similarities and differences that they see and eventually build up a model of an ancestral language that today‘s language group likely derived from.

The National Science Foundation and Summer Institute of Linguistics International have awarded more than $275,000 to create the two-year project. An earlier version of the program was developed in 2003 and tested by faculty and graduate students in the UH Linguistics Department, and by linguists in China and Indonesia.

The production version, Wordcorr 2, is available free to any linguist by visiting the Web site at