There’s an anecdote about Claude Shannon that describes the moment when the engineer was about to make an after-dinner speech at a symposium on information theory. Introduced as ‘one of the greatest scientific minds of our time’ the thin, white-haired man in his late sixties rose to his feet to tumultuous applause. The din of approval was so great that he could hardly make his voice heard. Finding the scenario ‘ridiculous’, Shannon took three balls from his pocket and, instead of delivering his address, entertained the guests with a juggling act. ‘It was as if Newton had showed up at a physics conference’, said the organiser. He was right. That is if Isaac Newton had published in 1948 A Mathematical Theory of Communication, dubbed the ‘Magna Carta of the Information Age’ by Scientific American.
The incident at the banquet was in 1985, more than a quarter of a century before Jimmy Soni and Rob Goodman wrote A Mind at Play, the first proper biography of the eccentric genius Shannon, whose intellectual prowess and scientific achievements are routinely ranked alongside those not just of Newton, but Albert Einstein. It’s a moment that so perfectly captures Shannon that the authors use it as their ‘cold open’ to introduce the life of a man distantly related to Thomas Edison who, like his illustrious forebear, contributed hugely to the technology template of the world today. As his obituary in MIT News states, Shannon was ‘the father of modern digital communications and information theory’, who in Soni and Goodman’s words was ‘the architect of the Information Age, whose insights stand behind every computer built, email sent, video streamed, and webpage loaded.’
Claude Elwood Shannon was born on the last day of April 1916 in the small town of Petosky on Lake Michigan in the Upper Midwest region of the US. Son of a businessman and a language teacher, the young Claude showed an early aptitude for technology while at Gaylord High School, where his best subjects were science and mathematics. His interest in communications manifested itself in the form of a half-mile barbed-wire telegraph system he constructed to connect with a friend’s house, as well as a part-time job as a messenger for Western Union. By 1932 he was studying electrical engineering and mathematics at the University of Michigan, where he first came into contact with the work of the nineteenth century English mathematician George Boole.
After graduating with two degrees he attended MIT, where he continued his studies in electrical engineering and first encountered Vannevar Bush’s mechanical analogue computer, which led Shannon to design switching circuits based on Boole’s work. His master’s thesis – A Symbolic Analysis of Relay and Switching Circuits (1940) – used Boolean algebra to establish the theoretical underpinnings of digital circuits. As his entry in Britannica states: ‘Because digital circuits are fundamental to the operation of modern computers and telecommunications equipment, this dissertation was called one of the most significant master’s theses of the 20th century. In contrast, his doctoral thesis, An Algebra for Theoretical Genetics (1940), was not as influential.’ In 1940 he went to the Institute for Advanced Study at Princeton, New Jersey where he was able to discuss his ideas with mathematicians such as John von Neumann, as well as the theoretical physicist Einstein.
Following his 1937 summer internship that was to inspire much of his subsequent research interests, Shannon joined the mathematical department at Bell Labs in 1941, and would remain affiliated with the industrial and scientific development company for more than three decades. In his early days there, during the Second World War he contributed to antiaircraft missile control systems and cryptography, which resulted in his post-war paper Communication Theory of Secrecy Systems, and which is credited with transforming cryptography from an art to a science. His work in cryptography brought him into contact with British mathematician Alan Turing who had been posted in Washington on an information-sharing initiative to discuss methods employed at Bletchley Park to break cyphers used by U-boats in the north Atlantic. The two met for tea and discussed Turing’s 1936 paper that defined the ‘universal Turing machine’. After the war, the National Defense Research Committee issued a summary of technical reports including an essay entitled Data Smoothing and Prediction in Fire-Control Systems that had been co-written by Shannon, and which addressed data and signal processing, effectively ushering in the Information Age.
I can visualise a time… when we will be to robots as dogs are to humans.
Claude Shannon (1916-2001)
Arguably the cornerstone of Shannon’s career, his 1948 paper A Mathematical Theory of Communication was published in two instalments in the Bell System Technical Journal. Building on the work of other Bell Labs researchers (including Swedish-American physicist and electronic engineer Harry Nyquist), Shannon’s masterpiece established the basic results of information theory in such a complete form that his framework and technology are still used today. In fact, the paper contains the first published use of the word ‘bit’ to describe a binary information digit; a term Shannon later credited to his colleague John Tukey, who also coined the term software for hardware-executable instructions, and had used the contraction in an internal Bell Labs memo the previous year. As one of his MIT colleagues Robert G. Gallager said: ‘Shannon was the person who saw that the binary digit was the fundamental element in all of communication. That was really his discovery, and from it the whole communications revolution has sprung.’
Unusually for an academic paper A Mathematical Theory of Communication has been cited tens of thousands of times, and was published in book form under the partially revised title of The Mathematical Theory of Communication with contributions by mathematician Warren Weaver. Although Weaver’s attempts to make the paper more accessible to the non-specialist worked to a degree, for the layman its contents are best summed up by Soni and Goodman who, in their eagerness to simplify the complex, perhaps oversimplify matters when they say that Shannon’s paper ‘explained how digital codes could allow us to compress and send any message with perfect accuracy.’
By the 1950s Shannon’s mind was engaged with machine learning. There’s a grainy film shot at Bell Labs in 1952 showing him standing alongside a 25-square maze that is solved by a robotic mouse called Theseus. Inspired by the tale of Theseus and the Minotaur from Greek mythology – in which the hero navigates his escape from the labyrinth by following a string guideline – the Shannon version shows how an electromechanical mouse ‘remembers’ its way across the metal floor to its target with the help of telephone relay switches. After much trial and error by the mouse, Shannon holds it aloft and tells the viewer that it has ‘finished learning’. It is then replaced, and the mouse completes the same task in a fraction of the time. Electrical engineer with Google Mazin Gilbert is in no doubt that Theseus ‘inspired the whole field of AI. This random trial and error is the foundation of artificial intelligence.’ The article ‘Bell Labs Advances Intelligence Networks’ agrees, stating ‘Shannon’s mouse appears to have been the first learning device of this level.’
Ultimately Shannon’s achievements rest on a handful of obscure articles. This may seem a harsh judgement on a scientific figure so widely revered, and yet it is one he’d almost certainly agree with. In a short piece called The Bandwagon (1956) he states that a ‘few first rate research papers are preferable to a large number that are poorly conceived or half-finished. The latter are no credit to their writers and a waste of time to their readers.’ Small wonder that he developed a reputation for being aloof and impatient; although his biographers defend Shannon by stating that he was simply on a different intellectual wavelength, which also meant that he didn’t waste time trying to win over his critics.
There is no doubt that Shannon was by any measure what might be regarded as eccentric. ‘There are plenty of mathematicians and engineers who write great papers’, say Soni and Goodman, but ‘there are fewer of them who, like Shannon, are also jugglers, unicyclists, gadgeteers, first-rate chess players, codebreakers, expert stock-pickers, and amateur poets.’ And while there is nothing strange about getting your pilot’s licence or learning to play jazz clarinet, the boundaries are pushed a little when we learn he installed a false wall in his house that rotated at the press of a button, and ‘once built a gadget whose only purpose when it was turned on was to open up, release a mechanical hand, and turn itself off.’
Claude Shannon died in 2001 at the age of 84 following several years with Alzheimer’s disease.
Poll finds engineers are Britain’s second most trusted profession
Interesting. Government ministers are nearly 50% more trusted than politicians! "politicians (11 per cent ), government ministers (15 per...