An Unconventional History of the Early IAS ComputerMarch 21, 2013
Why does Alan Turing (1912–1954) peer out from narrow slits in the dust jacket of a book about John von Neumann’s computer?
Why does Alan Turing (1912–1954) peer out from narrow slits in the dust jacket of a book about John von Neumann’s computer?
Turing's Cathedral: The Origins of the Digital Universe. By George Dyson, Pantheon Books, New York, 2012, 432 pages, $29.95.
Turing's Cathedral is an unconventional, surprisingly personal history. It centers on an early electronic computer, constructed at the Institute for Advanced Study in Princeton by a team led by John von Neumann. Planning for the machine began in 1946, and it was fully operational by 1952. It performed pioneering simulations of weather patterns, nuclear explosions, and what would later be called artificial life before being scrapped in 1958. Dyson knits many straggling narrative threads into this core story. He sketches the cast of characters involved with the machine and with IAS, including Klara von Neumann (John's second wife), Vladimir Zworykin of RCA (who tried and failed to provide memory tubes for the machine), legendary mathematician Kurt Gödel, visiting British computer pioneer Andrew Booth, and such Los Alamos scientists as Edward Teller and Stanislaw Ulam. Dyson expands these sketches to explore other work of his characters during the era; some of his story's threads originate in the prehistory of IAS, while others extend forward into a science fiction future of intelligent machines.
Why does mathematician, computer pioneer, code breaker, and posthumous gay icon Alan Turing (1912–1954) peer out from narrow slits in the dust jacket of a book about John von Neumann's computer? Dyson never explains his title directly. Perhaps it has something to do with the frenzy surrounding the Turing centennial in 2012. Publishers do like to sell books, and poor old von Neumann has faded almost completely from popular memory, while Turing's cultural prominence only rises. Turing does eventually appear in chapter 13, toward the end of the book, and a little later we unexpectedly discover that the titular cathedral is Google's corporate headquarters.
Dyson (son of the physicist Freeman Dyson) brings these characters to life more effectively than any other writer on the history of early computing. He begins by recalling a childhood spent roaming the halls of IAS to chat with its staff, and later mentions playing games with parts of the scrapped computer. His affection for both is palpable. The human characters are drawn with confidence. Thanks to his perseverance and personal connections, Dyson was able to track down and interview most of the surviving participants, including the reclusive Julian Bigelow, the leading engineer on the project; with well-chosen and often poignant quotations, he has encapsulated their motivations and struggles. He also obtained access to important papers from Klara von Neumann, which are still held in the basement of von Neumann's daughter Marina von Neumann Whitman. Dyson reconstructs Klara's story in compelling detail, rescuing an important contribution to early computing that was merely noted in passing by earlier authors. (Disclaimer: He has been kind enough to pass some of these papers along to me for my own research.)
The computer is similarly humanized. Dyson highlights the huge engineering challenges that faced all computer projects of the 1940s, and makes effective use of the machine's log book to dramatize the constant battle involved in keeping it running with tolerable reliability. One fortunate result of his efforts is that the computer project archives at IAS, where he spent time as a fellow researching the book, have been properly organized and made available to the public.
The tone and structure of Turing's Cathedral reminded me less of conventional history and more of the kind of nonfiction that bridges travel and science writing, practiced in different forms by such writers as Bill Bryson and John McPhee. Readers are collected at the beginning of a journey and deposited at the end, but in between their guide ventures down some unexpected byways, throws in long passages on tangentially related topics, and builds up to some personal musings on the meaning of life. If the author has a knack for weaving scraps together, and a voice that entertains rather than grates, the reader will leave the ride happier and wiser. I found Dyson's zesty prose engaging but would have preferred a tighter focus. Readers are tested early, as the world of early computing sketched in the first chapter abruptly vanishes. Von Neumann reappears eventually, but not before Dyson has led us though the entire history of Princeton, from the Lenape clans to the building of a permanent home for IAS.
I particularly welcome Dyson's explorations of the machine's applications to various areas of science, in particular its close relation with Los Alamos and the nuclear weapons program. Discussion of early machines has too often focused on their design and construction, at the expense of the work for which they were constructed. Dyson's book enjoys a far higher profile than any other recent work on the history of computing, publicized with a spate of media appearances and reviewed, seemingly, in every newspaper and magazine that still employs book reviewers. Many of its readers will know nothing of the history of computing, and will be immersed in a fascinating and richly depicted slice of history.
Yet, as a professionally trained historian, I very much hope that Dyson's idiosyncratic work will serve more often as an introduction to scholarly writing on the topic than as a substitute for it. His insistence on the IAS computer and its "fully electronic random access storage matrix" as "as close to a point source" for the origin of the "digital universe" as "any approximation can get" reflects an urge to explain a particular episode as the singular origin of something vast. Pinpointing beginnings is a primal driver of storytelling---consider the Book of Genesis. But historians have spent decades trying to move beyond partisan advocacy for one or another great man as the true inventor of the computer. Looking for a point source leads to history as viewed through a fisheye lens.
When discussing the influence of von Neumann on computing, historians traditionally focus on the 1945 "First Draft Report on the EDVAC" circulated under his name. His personal responsibility for many of the ideas set forth in the document has frequently been disputed, but its huge influence on the computer projects initiated over the next few years has not. Historians who have looked more closely at the era also credit an early description of the planned design for the IAS computer, circulated in 1946, and its early revisions as an important influence on many of these projects. The physical, functional computer was much less influential, in part because engineering delays resulted in its completion only after at least one of the machines modeled on its detailed design was already operational. One of Dyson's idiosyncrasies is to write as if these three achievements could not be separated, commenting relatively little on the 1945 "First Draft." He places the full burden of universe-changing historical importance on the physical IAS computer, which ran its first program in 1951, rather than on ideas that many others had already embraced, and indeed extended, years earlier.
Dyson boosts the historical importance of the IAS computer by omitting or downplaying information on developments elsewhere before or during 1951, with the exception of a tiny 1948 prototype computer at the University of Manchester from which von Neumann's team took the memory technology. Dyson's evidence is truthful, but startlingly incomplete. For example, while he concedes in his introduction that "the IAS machine was not the first computer," he never mentions EDSAC, operational at the University of Cambridge in 1949, which historians have almost universally recognized as the first useful computer built on the model described by von Neumann in 1945. It was also in 1949 that the Manchester team got its memory technology working in a full-scale computer. In 1951 UNIVAC provided the first commercially manufactured computer to the U.S. Census Bureau, while in the U.K. J. Lyons and Company, best known for its chain of teashops, completed its own computer and applied it to business automation. Other computers were already operational in the Soviet Union and Australia. All those milestones pass unmentioned by Dyson.
Dyson asks in conclusion, "How did the von Neumann vector manage to outdistance all the other groups trying to build a practical implementation of Turing's Universal Machine in 1946?" Even an attentive reader might assume that this "outdistancing" involved winning a race rather than losing it by several years. Dyson's implication that the various teams of computer builders inspired by von Neumann's proposed design all saw themselves as trying to implement the computational model described by Turing in his now-celebrated 1936 paper "On Computable Numbers, with an Application to the Entscheidungsproblem" is also likely to raise howls of protest from historians who have looked at early computing.
This disengagement with the historical literature limits the scholarly contribution of the book in other ways. For example, William Aspray's impeccably researched John von Neumann and the Origins of Modern Computing is acknowledged in the foreword but never cited in the body of the book; when the two diverge, it is hard to know whether Dyson considered Aspray's interpretation and rejected it on new evidence. Here is one example: In 1972 Nick Metropolis, a collaborator of von Neumann's and a pioneer writer on computer history, gave a well-known paper, "A Trilogy of Errors in the History of Computing." One of these common errors was calling von Neumann's computer "MANIAC." Metropolis insisted that this was not, and never had been, its name and that "MANIAC" should be reserved for the computer he had built at Los Alamos patterned after von Neumann's design. Aspray, and others who subsequently discussed the computer, therefore called it simply the "IAS computer." Alone among its cohort, it is denied a short, unique acronym. To Dyson, however, it is MANIAC throughout. This might be right. He may have learned the truth from new documents or from his oral histories. Perhaps the machine was known within von Neumann's team as MANIAC until someone realized that this glibness would not help their already rocky relationship with the pure scientists of IAS. Dyson does write that the name was "removed" by 1954. Perhaps Metropolis had a selective memory. But because Dyson ignored the existing consensus, rather than acknowledging it and justifying his challenge, he is unlikely to overturn it.
Dyson sometimes makes effective use of technical details to convey the world of early machines, as in his opening claim that in 1953 the world held just 53 KB of operational "high speed random access memory" (by which he means cathode ray tube storage), 5 KB of which were at IAS. At other points his reliance on metaphor rather than detailed explanation can be frustrating. Dyson puts great importance on the transition of computer memory from the one-dimensional tape in the conceptual Turing machine to the "two-dimensional" storage matrix found in a computer memory tube. Yet the physical tube existed in three dimensions, while its storage was treated by the programmer as a single series of numbered locations. The relevant shift was from sequential to random access, not from one- to two-dimensional storage.
The book is generously footnoted, and Dyson has supplemented his oral history interviews and newly discovered sources with visits to several existing archives. Yet his use of these sources is a little erratic. I am currently researching the Monte Carlo calculations performed by von Neumann's group on ENIAC in 1948 and had made copies of several documents cited by Dyson. Within his three pages on this topic and their footnotes I discovered eight distinct errors, mostly transcription problems in quotations, such as "16 cycles" for "160 cycles," omission of words in quoted passages without ellipsis, and the attribution of details from a 1949 letter to an episode in 1948. To Dyson's credit, he has made efforts to correct these in the forthcoming paperback edition. Those minor errors do not undermine the substance of his story, but their density makes me reluctant to rely on details elsewhere in the book without external confirmation.
The best way to decide whether you would enjoy Turing's Cathedral may be to check the reader comments at Amazon.com. Scores currently have bimodal distribution, with five- and two-star reviews the most numerous. Try to figure out which score might predict your reaction. The two-star reviews come from people frustrated by the book's insular perspective, insistence on the IAS computer as the singular origin point of modern computing, relaxed sense of chronology, lengthy digressions, and philosophical musings on the "digital universe." His fans praise Dyson's lively and elegant prose, eye for interesting details, and boldness in building what he himself calls a "creation myth" for the modern world. Both judgments are amply supported by the text.
Thomas Haigh (email@example.com) is an associate professor in the School of Information Studies at the University of Wisconsin---Milwaukee.