Chomsky, Wexler on the passing of Lila Gleitman (1929-2021)

Honoring Prof. Lila Gleitman of the University of Pennsylvania

August 16, 2021. Whamit! The Weekly Newsletter of MIT Linguistics.

Last week, we received the unspeakably sad news that Prof. Lila Gleitman of the University of Pennsylvania had passed away at the age of 91. An obituary article by her daughter Claire can be found reposted here by Barbara Partee (with whom she was collaborating on a paper up to the very last days).

Though Lila was never a member of our department, she was one of our most important colleagues nonetheless, a constant presence in our intellectual lives. From the very beginning, Lila was a close associate of the founders of linguistics at MIT and a guiding spirit for us all: both for the brand-new questions she asked again and again (urgent questions closely linked to the new turn that linguistics was taking when she began her career) and for the answers to these questions that she uncovered, in a lifetime of breathtakingly original studies of language acquisition — always presented with inimitable wit, energy, and charm.

Each of her numerous visits to MIT for talks and conferences was an occasion never to be forgotten; and she worked wonders as an active member of our departmental Visiting Committee from 2006 through 2014. She visited us for the last time, we think, as an invited speaker at a symposium honoring our colleague Ken Wexler on the occasion of his retirement in 2016, where the photo below was taken.

We asked two of our colleagues most closely associated with Lila and her work to share their thoughts on her passing: Noam Chomsky and Ken Wexler. Their remarks follow below.

from Noam Chomsky (Institute Professor Emeritus)

When a great scientist leaves us, the temptation is strong to say that she was a giant in her field. That doesn’t quite work for Lila Gleitman. The study of language acquisition, her primary scientific concern, was her field in a special sense: she virtually created the field in its modern form and led in its impressive development ever since.

Others had studied child language, but Lila was really the first to raise probing questions about what children know and how that knowledge is attained, questions informed by theoretical understanding and directed to issues of far-reaching importance. More than that, Lila developed elegant experiments to answer these questions, devising models that have guided experimental research throughout the years.

Many of the conclusions that Lila firmly established are quite startling. She and others working under her guidance were able to show that the child attains rich knowledge on the basis of very slight data, far beyond what the very young child exhibits in performance, suggesting that the data serve in fundamental ways as a triggering device. Her research also elucidates the means by which this remarkable achievement takes place.

Lila’s work constitutes a fundamental and highly original contribution not only to the study of language and its acquisition, but to the understanding of the human mind and its unique capacities far more generally.

Another reason why it is hard for me to write about Lila’s leaving us is that she was a close personal friend for 60 years. But those feelings hold for just about everyone who was lucky enough to have known her. Lila was one of those rare people who becomes a close friend virtually on first contact — an immediate reaction to her warmth, her wit, her sympathetic understanding. These qualities persisted unchanged, even amplified, through adversity that seemed only to heighten her rare personal qualities. It made her seem immortal.

And in important ways, she is.

-Noam Chomsky

from Kenneth Wexler (Professor Emeritus of Brain & Cognitive Sciences and Linguistics)

Lila Gleitman died a week ago Sunday, a sad event for everybody who knew her as well as our field. Lila had an extremely distinguished career in the study of language. It is not easy to think of a more distinguished psycholinguist. Her work was always characterized by an intense focus on careful and telling empirical analysis of fundamental issues. She never forgot her linguistic roots, and this interaction helped to mold her career’s work.

Her career is full of probing experimental questions on the most important issues. Who can forget her beautiful experimental take-down of the famous claims of Amos Tversky and his colleagues that seemingly precise terms were actually represented in the human mind in a much more approximate way, with all sorts of strange results. This extended all the way to the number system. “Which numeral is the most odd?” was an answerable experimental question in the view she was arguing against, a view that was beginning to dominate the field of psycholinguistics and cognitive psychology at the time. Lila knew that how you asked a question was as important as the statistical analysis of results.

When I searched the literature for useable studies of the linguistic input to the child developing language for my 1980 book with Peter Culicover, the most useful and careful one I could find was the beautiful empirical demonstration by Elissa Newport, Henry Gleitman and Lila that the input didn’t include a kind of “teaching language,” contra to the standard psychological claim at the time. This helped to establish the need for the child to have some prior knowledge of the nature of language. She pushed the view that the input to the child was extensively from surface information, reflecting, I believe, her studies with Harris with its emphases on distributional analysis. She produced quite a few empirical demonstrations of how this input could be used. In a beautifully simple experiment, she showed that even adults (and thus presumably children) couldn’t figure out the meaning of a verb from context, although they could for nouns. In Noam Chomsky’s terms the meaning of a verb is then not “epistemologically available.”

My comments barely scratch the surface of Lila’s work. She produced generations of students who filled psychology departments around the country and who were strongly devoted to her. When we first met we hit it off instantly, becoming good friends. I will miss her and her constant interest, ideas, conversations, work and general support for our great enterprise. Who will be that thinking psychologist who allows for the hope of integration of our fields?

-Ken Wexler