Unless you spent last year living in a cave, it was more or less impossible not to know 2012 was the Turing Centenary year. While it was gratifying to see so many exciting events arranged in Turing's honor, I could not help thinking we had moved more or less seamlessly from a position where Turing was largely ignored, to one where it sometimes felt as if no one else counted. In his editorial for the January 2013 Communications,5 Moshe Vardi similarly gives voice to a very reasonable concern that "we may have gone from celebration to hagiography."
Maxwell (Max) Herman Alexander Newman, whose own centenary passed relatively quietly in 1997,a was closely associated with Turing,3 and from the mid-1930s onward played an important part in promoting and shaping Turing's career and promulgating as well as implementing his ideas on computing.b
Despite being largely unknown in many computing circles, Max Newman was, in his own right, one of the most significant figures in the early history of British computing. He was active in the field for more than 10 years initially at Cambridge before World War II, then at Bletchley Park during hostilities, and finally in the postwar setting of the University of Manchester in the mid-late 1940s. Newman's very English habit of understating his personal contribution, combined with his preference for stressing the accomplishments of others may explain why historians of computing have generally paid little attention to this extraordinary man who is, in consequence, princi- pally remembered as a mathematician working in the field of topology.
We have heard claims over the last year that modern computing began with Alan Turing. However, the roots of computing run much deeper, and any "starting point" is open to being contested. We can, for example, trace an interesting path from Turing back through Newman to a lecture delivered by David Hilbert at the Sorbonne at the turn of the 20th century, in which he proposed 23 "future problems." These were taken up by the mathematics community and collectively formed the agenda for mathematics research in the century that followed. Hilbert's 10th problem can be framed as asking if there could exist, in principle, a definite process involving a finite number of steps, by which the validity of any given first-order logic statement might be decided? Turing first met Hilbert's so-called Entscheidungsproblem around the spring of 1935 when he was a student in Newman's Part III Foundations of Mathematics course. Newman later recalled: "I believe it all started because he attended a lecture of mine on foundations of mathematics and logic in which I had mentioned...that what is meant by saying that the process is constructive is that it's purely... mechanical...and I may even have said a machine can do it. But he took the notion...and did produce this extraordinary definition of a perfectly general,...computable, function thus giving the first idea really of a perfectly general computing machine."2
Had Newman done nothing else in his career, Colossus would have entitled him to a place in the pantheon of pioneers of early British computing.
In April 1936, Turing gave Newman a draft copy of his dazzlingly original answer to the Entscheidungsproblem, at the center of which was an idealized description of a person carrying out numerical computation that, following Church, we have come to call a "Turing machine." All modern computers are instantiations of Turing machines in consequence of which Turing's paper is often claimed to be the single most important in the history of computing.
Newman almost immediately took Turing under his wing, canvassing successfully for "On Computable Numbers" to be published by the London Mathematical Society. At the same time he enlisted Alonzo Church's help in arranging that Turing should spend some time studying in Princeton.
Cambridge in the late 1930s and early 1940s was particularly fertile soil for computing pioneers, and Newman played a part in the education of most of them. In addition to Alan Turing and his contemporary Maurice Wilkes, other students of Newman's included Tom Kilburn, Geoff Tootill, and David Rees.
In 1939, as war was spreading across Europe, Newman was awarded a fellowship of the Royal Society. However, he had little opportunity to use this fellowship to provide impetus for further work in mathematics, as the outbreak of hostilities took one colleague after another out of academic life into war work. Newman, whose young family had evacuated to the U.S., grew increasingly isolated and slowly became disillusioned with life at Cambridge. Acting at the suggestion of Patrick Blackett, he accepted a post at Bletchley Park. Neither of them could have had the least notion that Newman had thereby embarked on a course that was to completely alter the future direction of his career. Newman initially began work as a cryptanalyst working on "Fish"c traffic as part of John Tiltman's group. The type of transmission that attracted the greatest interest was known as "Tunny" and carried messages between the very highest ranks of the German command. Manual methods utilizing statistical techniques had been devised for breaking into the code, but the sheer volume of traffic being intercepted was beginning to overwhelm the human resources available.
Newman was able to make a decisive contribution. He believed it was possible to mechanize the attack on Tunny and lobbied successfully to test his conviction by developing an electromechanical code-breaking machine that came to be known as the "Heath Robinson." Newman was placed in charge of the development, and a cryptanalyst, two engineers, and 16 Wrensd were placed at his disposal. He ran his section which, following the Bletchley Park tradition of naming sections after those in charge of them, was called "The Newmanry," in a democratic spirit, encouraging his staff to speak up if they thought him mistaken. Democracy notwithstanding, Newman was later recalled by Donald Michiee as leading with patriarchal authority and having something of the force of nature about him. Possessed of considerable intellectual daring, he demanded, and generally got, the impossible both of events and people. Michie further recalls that Newman proceeded with "vigor and certitude, seemingly as a vehicle without reverse gear." Under Newman's leadership originality flourished. The successes of his team, in which he took great pleasure, were not the result of detailed micromanagement but came about by finding people in whom he could place his trust and allowing them to work according to their own judgment. This allowed Newman to concentrate on the broader managerial and organizational issues in the service of which, in Michie's words, he displayed an "unerring sense of direction in a broader-brush landscape of which [his staff] often had no inkling. Over time his persistence toward a perceived goal would fructify in a stunning coup."
Although the Heath Robinson proved fairly unreliable, the results it achieved were sufficiently encouraging for approval to be granted to develop a more sophisticated machinethe Colossus. A great deal has been written about this machine, the world's first digital electronic computing device. Suffice it to note that, had Newman done nothing else in his career, Colossus would have entitled him to a place in the pantheon of pioneers of early British computing.
As it was, the Colossus profoundly affected Newman's future career. He saw at once, as few others did, the impact computing would come to have on mathematics and he decided to establish a computer-building project as soon as the war was over. In Newman's judgment, the mathematics department at Cambridge was not the right environment for this, and he began to look around for a more suitable setting. With Blackett's help and encouragement, Newman was appointed to the Fielden Chair of Pure Mathematics at Manchester University, the post having become vacant when Louis Mordell moved in the opposite direction to take up the Sadleirian Chair. Newman had two clear goals in mind: To establish a first-rate department that could stand comparison with the best in the country and to build a computer. At Bletchley Park, Max had been surrounded by people who could help him achieve both objectives.
In a clear declaration of intent, Newman brought with him Jack Good and David Rees, both of whom had been at Cambridge before the war, and as part of Newman's section at Bletchley Park, they had experience working on Colossus. With further assistance from Blackett, a substantial grant was obtained from the Royal Society. This was an innovative use of Royal Society funds and was the first award made for the purpose of developing a computer. The only piece of the puzzle that was missing was a lead engineer. Newman was not the only person looking for a top-flight engineer: the National Physical Laboratory (NPL) was also planning to build a computer and "good circuit" men, Newman wrote to von Neumann, were "both rare and not procurable when found."4
Thus it was that, at war's end, F.C. (Freddie) Williams found himself in the fortunate position of being a man much in demand. It must have been quite a disappointment to the NPL when, in late 1946, Williams was offered and accepted the Edward Stocks Massey chair of Electro-Technics at the University of Manchester. Blackett and Newman, who had earlier discussed with I.J. Good the possibility of hiring Williams, were both on the appointments panel.
Having secured the support of the university, obtained funding from the Royal Society, and assembled a first-rate team of mathematicians and engineers, Newman now had all elements of his computer-building plan in place. Adopting the approach he had used so effectively at Bletchley Park, Newman set his people loose on the detailed work while he concentrated on orchestrating the endeavor. The result was success beyond all expectation. By the middle of 1948 the Small Scale Electronic Machine (SSEM) was up and running. Although little more than a proof of concept, it was still the world's first working digital electronic stored program computer. The Manchester team had indisputably achieved a major coup.
Newman's direct involvement with computing activity was, however, coming to an end. Like Blackett, Newman was opposed to the inevitable use of the Manchester computer in the development of nuclear weapons, and as the government took an ever-closer interest in the Manchester computer, Max stepped back to leave further development to the engineers.
Newman retired in 1964 but continued to be active in topology and two years later produced an engulfing theorem for topological manifolds. Until 1970, he taught in a succession of visiting professorships, mostly in the U.S. In May 1973, his wife Lynn died and later the same year he married Margaret Penrose. Together they enjoyed a happy and contented life surrounded by their children and occupied with travel, music, and entertaining. Newman died in 1984 at age 89.
Newman's influence on the first generation of British computer scientists was incalculable.
Newman was a deeply cultured man with an inquiring mind whose interests ranged over a broad spectrum. His influence on the first generation of British computer scientists was incalculable, and his early appreciation of the importance of computing was probably matched only by that of Alan Turing. The vision and leadership he showed at Bletchley Park during World War II and his single-minded determination to mechanize the British code-breaking efforts not only had an appreciable impact on the outcome of the conflict but also created a computing legacy that he was determined to carry into the postwar situation. Such was the deftness by which he accomplished the transfer of knowledge that some of those who gained most from his understanding were more or less unaware of the singular contribution made to their own success by this remarkable man.
I am content to leave the last word to Mary Cartwright, who in presenting Newman with the 1962 De Morgan Medal said of him: "Newman is a scholar of mathematics with a fine sense of the directions in which important advances are to be expected. It was, for instance, at least in part as a result of his interest and advocacy that Manchester University acquired one of the first general-purpose computers. He has also contributed greatly to mathematics as an expositor....Whether in writing or in speaking he has a distinction and clarity of language that few can rival and all must admire."1
4. Newman, M.H.A. Letter to John von Neumann. In Box 6, Folder 2, Item 2, The Newman Digital Archive, The Max Newman Digital Archive, the University of Portsmouth Future Proof Computing Group, and St. John's College, Cambridge (Feb. 8, 1946) (unpublished letter).
b. For a much fuller account of Newman's contribution to the early development of computers in the U.K., see Anderson, D. "Was the Manchester 'Baby' conceived at Bletchley Park?," BCS eWIC, (2008); http://www.bcs.org/upload/pdf/ewic_tur04_paper3.pdf.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2013 ACM, Inc.
No entries found