He Wanted Out of Grad School, So He Wrote His Dissertation in Five Weeks
With no thesis and a June deadline, Paul Graham stitched together his unfinished book on Lisp and called it a dissertation.
paulgraham.com · What I Worked OnIn April 1990, Paul Graham, then a computer science doctoral student at Harvard, ran into his adviser, professor Tom Cheatham, and was asked a simple question. Was he nearly ready to finish in time for a June graduation?
He had not written a word of his dissertation. He was taking art classes at Harvard. He was planning to be an artist, and he was still working on his book about Lisp.
Those side projects had taken over. Like many graduate students, as Graham later put it, he "was working energetically on multiple projects that were not my thesis." For him that meant the art classes and work on Lisp, a programming language then popular with artificial intelligence researchers, which became his book On Lisp. Cheatham was easygoing and never commented on the strange classes, but the thesis remained unwritten.
Graham felt stuck. Quitting outright did not appeal to him, but he could not see how to keep going. He admitted he envied his friend Robert Morris, a fellow programmer who had been kicked out of Cornell for creating the 1988 internet worm, for finding a "spectacular" exit from graduate school.
So when Cheatham asked about June, Graham made a snap decision. In "what must have been the quickest bit of thinking in my life," as Graham later put it, he decided on the spot to try to write a dissertation in the five weeks or so before the deadline by reusing parts of his unfinished book wherever he could. With no perceptible delay, Graham answered, "Yes, I think so. I'll give you something to read in a few days."
He picked applications of continuations as his topic. A continuation is a way for a program to save what it should do next, so it can pause and resume that work later, a trick for controlling how software runs through complicated tasks. It was not the topic he cared about most. In retrospect, Graham wrote, "I should have written about macros and embedded languages," tools that let programmers reshape a language itself, because a whole world there had barely been explored.
He only needed to finish. He assembled the dissertation quickly from his book material, turned it in, and it passed, but barely. Five weeks of borrowed work closed out years of graduate school.
He was free to go learn to paint.
Questions people ask
What was On Lisp, the book he borrowed from?
On Lisp was Graham's technical book about Lisp, a programming language prized for letting programmers bend the language itself. It was published in 1993, after the dissertation episode, and much of the material he was writing as a graduate student went into both projects.
What are continuations, the topic he chose?
A continuation captures what a program should do next, so the program can pause, save that future, and resume it later. Programmers use them to manage backtracking, web requests, and other jobs where control jumps around. They are a classic but tricky idea in programming languages.
What happened to Robert Morris after the 1988 worm?
Morris released what became known as the Morris worm in 1988 while he was a graduate student at Cornell. It spread widely across the early internet and led to his conviction under computer fraud law. He later became a professor at MIT and a co-founder of the startup Viaweb with Graham.