Claude Elwood Shannon (April 30, 1916–February 24, 2001) was an American mathematician and electrical engineer whose work established information theory and helped provide the mathematical foundations of digital communication and computing. His research connected Boolean algebra with switching circuits, quantified information and the limits of reliable transmission, and developed mathematical approaches to cryptography. He worked at Bell Laboratories and later held a professorship at the Massachusetts Institute of Technology (MIT). (news.mit.edu)
Education and academic career
Shannon was born in Petoskey, Michigan, and grew up in Gaylord. In 1936, he received bachelor’s degrees in mathematics and electrical engineering from the University of Michigan before entering graduate study at MIT. He received his master’s degree in electrical engineering and doctorate in mathematics in 1940. His doctoral dissertation concerned an algebraic treatment of population genetics, a subject distinct from the communication research for which he became best known. (news.mit.edu)
He joined Bell Laboratories in 1941. During World War II, his research included military fire-control problems and secrecy systems. He became a visiting professor at MIT in 1956 and served as Donner Professor of Science from 1958 until becoming professor emeritus in 1978. His work combined abstract mathematical analysis with practical questions about circuits, signals, and machines. (news.mit.edu)
Boolean algebra and switching circuits
Shannon’s early work showed how algebra could describe the operation of electrical relays and switches. His paper “A Symbolic Analysis of Relay and Switching Circuits,” published in 1938 from his master’s research, treated circuit conditions symbolically rather than relying solely on diagrams or trial-and-error design. The approach connected electrical switching with logic: combinations of switches could implement logical operations, and algebraic transformations could simplify the corresponding circuits. (tubes.mit.edu)
For switches represented by conducting and nonconducting states, series and parallel arrangements express different combinations of conditions. Shannon developed methods for analyzing these arrangements and constructing circuits that satisfied specified requirements. This made it possible to move systematically between a logical expression and its physical realization. The contribution was a general design method, not the invention of the computer itself; it helped establish the theoretical basis of digital switching systems. (tubes.mit.edu)
Information and communication
Shannon’s “A Mathematical Theory of Communication” appeared in two installments in the Bell System Technical Journal in July and October 1948. Building on earlier communication research, it incorporated noise and the statistical structure of messages into a general mathematical framework. His model distinguished an information source, transmitter, channel, receiver, and destination, with noise potentially altering the transmitted signal. (princeton.edu)
The theory deliberately separated communication engineering from semantic meaning. Its question was how to reproduce a selected message, not whether that message was true, useful, or meaningful. For a discrete random variable (X) with probabilities (p_i), Shannon defined entropy as
[ H(X)=-\sum_i p_i\log_2 p_i. ]
With base-two logarithms, entropy is measured in bits. It expresses average uncertainty: equally likely alternatives produce greater entropy than alternatives dominated by one highly probable outcome. (princeton.edu)
The source coding theorem established limits on lossless compression. The noisy-channel coding theorem showed that suitable coding can make transmission errors arbitrarily unlikely at rates below channel capacity, under the theorem’s assumptions. These are asymptotic existence results; they do not imply that every finite error-correcting code achieves capacity or eliminates errors completely. (princeton.edu)
Mathematical cryptography
In “Communication Theory of Secrecy Systems,” published in October 1949, Shannon applied probabilistic reasoning to encryption. The paper drew on a classified report completed in September 1945 and examined the mathematical relationships among messages, keys, and encrypted texts. (onlinelibrary.wiley.com)
He formalized perfect secrecy: observing the ciphertext leaves an observer’s probabilities for the possible plaintexts unchanged. A correctly used one-time pad, with an independent, uniformly random key used only once, satisfies this criterion. Perfect secrecy concerns information available to an observer, rather than the computational difficulty of breaking a cipher. Shannon also discussed confusion and diffusion as principles for obscuring relationships between keys, plaintexts, and ciphertexts. His analysis distinguished fundamental security limits from the practical properties of particular encryption systems. (cs.virginia.edu)
Chess and artificial intelligence
Shannon’s 1950 paper “Programming a Computer for Playing Chess” examined how a machine could select moves without exhaustively considering every possible game. It combined search through a game tree with a heuristic evaluation function and minimax reasoning. He distinguished a Type A strategy, examining variations to a fixed depth, from more selective Type B strategies. Chess served as a precisely defined experimental problem for investigating machine decision-making. (cs.cornell.edu)
On August 31, 1955, Shannon coauthored the proposal for the following summer’s Dartmouth research project with John McCarthy, Marvin Minsky, and Nathaniel Rochester. The proposal helped establish artificial intelligence as a named research field and included questions about learning, abstraction, language, and self-improvement in machines. (www-formal.stanford.edu)
Experimental machines and recognition
Shannon also constructed electromechanical devices. Theseus, created in 1950, used a magnetic mouse controlled by relay circuitry beneath a configurable maze. The apparatus searched for a target and retained information that enabled subsequent navigation. Its apparent learning depended on the external control mechanism, rather than electronics contained within the mouse. (news.mit.edu)
Other creations included mechanical juggling devices and recreational machines built in his home workshop. A collection of these objects was donated to the MIT Museum in 2007. Shannon received the National Medal of Science and IEEE Medal of Honor in 1966, and the Kyoto Prize in Basic Sciences in 1985. He died in Medford, Massachusetts, aged 84. (news.mit.edu)