Michael Kenstowicz

Dear Noam,

Last year I reread The Sound Pattern of English in connection with writing the chapter on SPE for Elan Dresher and Harry van der Hulst’s forthcoming Handbook on the History of Phonology. It brought back fond memories of grappling with this difficult book as a graduate student at the University of Illinois, where I was mentored by your first generation of MIT graduates: Bob Lees, Ted Lightner, Arnold Zwicky, and Mike Geis. I thank you and Morris for creating a field to which I along with so many others have devoted my life. For your interest, I reproduce below the closing passage from my chapter in Elan and Harry’s volume.

In the fifty or so years since its publication no single work has come close to SPE’s combination of theoretical originality and descriptive insight. Anyone studying some aspect of the stress or segmental morphophonology of English today first looks to see what SPE might have said about the matter. Most of the major issues facing the field today were addressed by SPE with specific proposals and clearly articulated positions. Questions about data vs. theory, gradience vs. abstraction, competence vs. performance, etc. that were thought settled (at least within the field of generative linguistics) are now being raised anew as researchers adopt experimental probes of linguistic competence and computational modeling of grammar and its acquisition inspired by machine learning techniques over large corpora of speech and text. Time will tell whether these approaches yield insights comparable to those achieved by the SPE program initiated by Chomsky and Halle in the middle of the last century.