I was walking around a 1959 issue of the Proceedings of the Institute of Radio Engineers I found this fantastic article by the great Walter Pitts. What was Mr. Pitts and a frog’s brain article doing here? Well, it all makes sense, really, and I was very happy to unexpectedly find Pitts here when I was looking for something else. Walter Pitts (1923-1969) is one of those very important but now somewhat faded names in the history of American science. He was a brilliant young man turned into a brilliant mature one, a logician/cognitive psychologist who corresponded with Russell on the Principia at age 15 and who went on to develop (with Warren McCulloch) some of the fundamental ideas in what would be known as neural networks. The work was really much more far reaching, touching work in the fields of artificial intelligence, cybernetics, neuroscience(s), philosophy, cognitive sciences and of course computer science.
Pitts was a genius I’m sure, a sort of Matt Damon-type character from the movie Good Will Hunting1. He was offered a chance to study with Russell in the U.K. at 15, but refused and wound up running away from home to go to the University of Chicago to listen to Russell lecture, and where he stay on, finding another great logician, Rudolf Carnap (then at Chicago) and offering him critiques on his new work. Pitts would blow like the wind in and out of Carnap’s office, the logician spending months trying to find out who and where Pitts was–and what he was was a homeless teenage boy. Once found, Carnap got Pitts a low-level maintenance job. I’m not sure what Pitts was doing before he switched ships and jumped (still homeless) to MIT in 1942, where he was introduced and was impressive to the not-to-be-impressed Norbert Wiener, and where he also met Warren McCulloch, with whom he would work on the Leinbizian idea of the nervous system as a sort of universal computer, publishing their outstanding “A Logical Calculus of Ideas Immanent in Nervous Activity” in September 1943.2
Right before this paper–which is considered to be a founding effort in the field of neural networks–Pitts published three other papers of interest in this same area (and in the same journal)–all published when he was 19 years old. “Some observation on the simple neuron circuit”, “The linear theory of neuron networks: The static problem”, and “Some Observations on the Simple Neuron Circuit” were all efforts leading up to his paper with McCulloch one year later. Impressive work, especially for a 17-year-old, and particularly for a new, high entry point journal published by the University of Chicago, entertaining the logical structure of neural nets as coded circuitry, a new way of looking at the mechanistic view of philosophy of mind.
Things however really didn’t go well for Pitts: he was fiercely, probably tragically “shy” or demanding of his own attention, exceptionally modest, and with very few personal relationships. Just a few years later, a rift developed between the wife of Norbert Wiener and McCulloch, and was of such vehemence that anyone maintaining a friendship with McCulloch would be shunned by Mrs. (and therefore, necessarily in this case, Dr. )Wiener. And thus Pitts remained friends with McCulloch and lost contact with just about everyone else, making his working/contributing life quite difficult. Little is known about what happens to Pitts in the 1960’s, but it wasn’t good, and he wound up dead, featuring symptoms indicative of severe alcohol abuse, in 1969. He was 46, and his best work was already 25 years old.
_______
Notes:
1. We read in this wonderful account of Pitts from Neil R. Smalheiser’s appreciation, Walter Pitts ( found in: Perspectives in Biology and Medicine – Volume 43, Number 2, Winter 2000, pp. 217-226):
[Introduction ] “The movie Forrest Gump made the point that the greatest, most heroic Americans are people of extraordinary character who flicker briefly into public consciousness and are quickly forgotten. Walter Pitts was pivotal in establishing the revolutionary notion of the brain as a computer, which was seminal in the development of computer design, cybernetics, artificial intelligence, and theoretical neuroscience. He was also a participant in a large number of key advances in 20th-century science. Yet while his contemporaries Alan Turing, Ludwig Wittgenstein, and John von Neumann entered the pantheon of fame, Pitts remains a shadowy folk hero. Stories about Pitts have circulated among the cognescenti for years, but almost nothing has been written about him. Here, I have collected reminiscences from his friends and associates to provide a unique insight into a remarkable life; if some exaggerations and embellishments have crept in, they only underscore the basic truth that Pitts was a man with Gumption ability…”
2. Warren McCulloch remembers Pitts in this way: “In 1941 I presented my notions on the flow of information through ranks of neurons to Rashevsky’s seminar in the Committee on Mathematical Biology of the University of Chicago and met Walter Pitts who then was about seventeen years old. He was working on a mathematical theory of learning and I was much impressed. He was interested in problems of circularity, how to handle regenerative nervous activity in closed loops. … For two years Walter and I worked on these problems whose solution depended upon modular mathematics of which I knew nothing, but Walter did. “ (McCulloch, 1989, pp. 35-36, cf. McCulloch, 1965a, pp. 9-10).
