When not in a classroom or laboratory, Maria Polinsky spends time doing fieldwork off the southeast coast of Africa, in Madagascar. She studies Malagasy, a melting-pot language whose influences start in Borneo and now borrow from Swahili, Arabic, and French.

Or you might find Harvard’s newest professor of linguistics in the Eurasian Caucasus, just south of her native Russia. She calls this region of mountains and lowlands “a linguist’s paradise” or “a mountain of tongues” because of its linguistic variety. Languages there belong to three separate linguistic families.

Polinsky, the first woman to hold a senior position in linguistics at Harvard, studies two of those families, and this spring taught a course in linguistic field methods.

She speaks Russian, English, and French, and gets along in Malagasy and in a string of lesser-known tongues from the Caucasus, including Tsez — with its 7,000 native speakers and its 120 cases.

That’s the most cases of any language. English has three. The case of a noun or pronoun signals its grammatical function. In English, for instance, “he, him, his” illustrate respectively the subjective, objective, and possessive cases.

Experts like Polinsky study language and its structure to identify the grammatical architecture behind spoken words. One of the big questions in linguistics: What makes languages the same, and what makes them different?

On the surface, languages seem very different, said Polinsky. English, for instance, has no genders, and German three. Still other languages require six to 16 genders to signal grammatical “agreement” between a noun and an article or a noun and a verb.

“But when we look at how [languages] are built,” said Polinsky, “there are a lot of similarities.” By looking at them, she said, “we get closer to understanding what it means to [speak] a human language.”

Linguistics is pulling back from its traditional roots in philology, said Polinsky — that is, from an association with literary, historical, and cultural branches of language study. Practitioners of the discipline have a great deal in common these days with computational scientists, psychologists, and biologists who specialize in the brain. Linguistics, she said, is in some ways more an engineering field than a cultural or artistic one.

At Harvard, Polinsky is part of the Mind/Brain/Behavior (MBB) Interfaculty Initiative, a cross-disciplinary group of investigators who share an interest in behavior and cognition as they relate to the human nervous system.

“The strength of the [MBB] program here was one of the things that attracted me to Harvard,” Polinsky said. “We’re at a stage when there’s going to be a lot of interdisciplinary research.”

Polinsky taught a course last fall in theoretical and experimental linguistics, an attempt “to bring together the salient theoretical questions of linguistics, and see how the evidence from mind and brain [science] can shed light on them,” she said, “and how we can build bridges between linguistics and cognitive science.”

Linguistics is allying itself with experimental sciences, but Polinsky is also investigating a question of direct social and cultural interest: What happens to a language learned in childhood, then abandoned later for a dominant language?

Linguistics scholars call such “abandoned” languages “heritage languages,” and the phenomenon is very common. In the immigrant-rich United States, said Polinsky, about a third of undergraduates entering college have a heritage language. But they are reluctant and even embarrassed to use it. So this language resource remains “untapped and underutilized,” she said.

Polinsky studies these reluctant speakers in a laboratory setting, measuring their behavioral responses to foreign language data, including the speed of pushing a button at the sound of prompts in the lost language.

On the horizon, there are more elaborate ways of measuring such responses, including brain mapping or eye movement. “But when you start working in language, you can get a lot of information from very simple behavioral responses,” said Polinsky. “Before you do something fancy, you want to do something simple and readily accessible.”

Implicit in her research is a social need to revive this vast pool of latent knowledge, she said — including getting the word out to educators and funding agencies. “What I do,” said Polinsky of her scientific heritage language studies, “is just a small fragment of what needs to be done.”

Relearning how to speak and write a foreign language would be a boon to the education of these students, she said, and a boost to understanding the cultures they came from. “It can only be better if you know both cultures,” said Polinsky. She’s writing a book on heritage languages, whose working title is “A Language Never Born.”

A multiuniversity team Polinsky is part of just got a five-year grant from the U.S. Department of Education to study heritage speakers, and to hold annual national conferences. (The first will be in August, at the University of California, Davis.)

Polinsky is the mother of two sons. Abi, 13, is in seventh grade. Lev, 28, graduated from Harvard College in 2000 and works for JPMorgan Chase & Co. in Manhattan.

In life outside linguistics, Polinsky claims to be “your basic boring academic,” and enjoys reading, the theater, and music. (She likes live jazz and classical music — and is “very proud to live in Cambridge” because Cold War-era singer and satirist Tom Lehrer still lives there.)

Polinsky came to Harvard last year by way of the University of California, San Diego, where she had taught and done research since 1996. The West Coast was her first residential experience with American culture. “California gives you a lot of inner space, where you can be by yourself and work,” she said.

Cambridge, on the other hand, “is like going back to Moscow,” said Polinsky, because of its colder climate and its engaged, conversation-centered culture. “It’s like going back to a language I haven’t spoken for a while.”