Skip to Content
MIT News magazine

Language Reclamation 101

How MIT linguists are working to revive Wôpanâak.

When the legendary MIT linguistics professor Ken Hale retired in 1999, associate professor Norvin Richards, PhD ‘97, a specialist in Native American and aboriginal Australian languages, was hired to replace the man he considers his mentor. Today, Richards continues Hale’s work on the Wôpanâak Language Reclamation Project by collaborating with Jessie Little Doe Baird, SM ‘00, and Nitana Hicks, SM ‘06, on the dictionary of this Native American language.

MIT associate professor Norvin Richards, PhD ’97, consults MIT’s 1685 edition of the Eliot Indian Bible as part of the Wôpanâak Language Reclamation Project.

Richards classifies the two main tasks of a linguist doing language reclamation as “dictionary work” and “grammar work.”

The “dictionary work” using the Eliot Indian Bible involves reading parts of the Bible in search of words that are not yet in the lexicon. “For example,” Richards explains, “I recently finished reading the Book of Isaiah. When I came to an unfamiliar word, I tried to figure out how it would have been pronounced and what it means. Sometimes discerning the meaning and the pronunciation of a new word is comparatively simple. This language (like the rest of its Algonquian relatives) has a lot of words that are composed of isolable smaller parts (which linguists call ‘morphemes’). For example, there’s a suffix -nuhsh that you can attach to a verb to get a new verb that means ‘do something on someone else’s behalf.’ By adding it to a verb that means ‘lift,’ you get a new verb that means ‘lift for someone.’ There are many other examples of suffixes (and, less commonly, prefixes) that can be added to verbs and nouns to change their meanings in various ways. Sometimes, when I find a new word, it’s really a combination of familiar elements: some verb that we already know but haven’t yet seen with (for example) the suffix -nuhsh.

“Sometimes, on the other hand, I come across a word that isn’t just a combination of familiar morphemes. In the Book of Isaiah I found a word for ‘nettle’ (that is, the stinging plant), which was spelled ‘masson’ in the text. When I find a word like that, I have to do some work to figure out how the word would have been pronounced. Wôpanâak has several vowels that are often hard to distinguish. … In order to narrow down further how the Wôpanâak word is pronounced, I do some etymology. One of the early discoveries of linguistics is that sound changes are regular. That is, when we have two related languages, it isn’t just that they have a bunch of words that sound similar; there are actual rules for how the sounds correspond with each other. … We rely on sound correspondences–laws like ‘If Abenaki has a z, then it must be an s in Wôpanâak’–which we’ve established by looking at clear cases where we’re sure what’s being spelled, and figuring out the rules for relating the Wôpanâak words to their counterparts in other languages.”

“ ‘Grammar work’ is harder to describe straightforwardly,” Richards continues, “but basically, it involves looking at the whole Bible for patterns. One of my first projects of this kind was something Ken suggested that I do when I was just getting started. Verbs in this language (as in most Algonquian languages) come in two major types (called conjunct and independent), depending on their use in the sentence. So, for example, verbs in relative clauses must use the conjunct form; questions asking ‘why’ are in the conjunct form, and so are most (but not all) questions asking ‘who,’ while questions asking ‘what’ are in the independent form. The rules conditioning the choice are very complicated. Since questions are one place where the conjunct form is sometimes used, I spent a lot of time looking through the Bible for questions, categorizing the verbs as conjunct or independent, and then trying to develop generalizations about when which form is appropriate.”

“Some kinds of things in the non-Biblical texts are more useful than the Bible,” Richards explains. “For one thing, the Bible tends to preserve the English word order, so if we’re trying to study word order, Native Writings in Massachusett, edited by Ives Goddard and Kathleen Bragdon (Philadelphia: American Philosophical Society, 1988), is generally more helpful. On the other hand, the Bible is a much larger text. The plan eventually is to go through all available texts. After the Bible, I want to read through another book by Eliot, The Logic Primer, published in 1672. It’s intended for use as a training manual for Wampanoag Christians who want to become ministers; it’s supposed to teach them how to construct logical arguments to help convert their people. It’s full of interesting words like ‘syllogism,’ ‘subject,’ and ‘predicate.’” Richards finds it convenient to study an electronic copy of the Eliot Indian Bible on the Web at Early English Books Online, but he periodically checks lists of unclearly scanned words and phrases against the print copy owned by MIT’s Archives and Special Collections, a gift from I. Austin Kelly, ‘26.

Richards first became passionate about endangered languages when he helped create a dictionary of Lardil on a trip to Australia with Hale in 1996. “Ken was a remarkable fieldworker for lots of reasons,” Richards says. “Fieldwork involves a lot of data, which flows past you in a fairly disorganized way. He had an amazing memory and ability to spot patterns, which probably had something to do with his legendary ability to learn languages. But I think one of the secrets of his success was his philosophy. He didn’t see himself as better or more important than anyone else. It’s because of him that Jessie was able to get the education that she got, and the type of partnership that Jessie and I have now is one of the kinds of things he worked his whole life to build.”

Read the related MIT News feature: “Saving a Language

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.