Richard Hamming
Richard Wesley Hamming (February 11, 1915 – January 7, 1998) was an American mathematician and computer scientist whose work in numerical analysis, coding theory, and digital signal processing profoundly influenced modern computing and communication. He is best known for inventing Hamming codes, one of the first families of error-correcting codes, and for introducing the Hamming distance metric. During his long career at Bell Labs (1946–1976), he also contributed to the Manhattan Project, developed numerical methods for scientists and engineers, and wrote influential books including *Numerical Methods for Scientists and Engineers* and *The Art of Doing Science and Engineering*. His 1986 talk "You and Your Research" remains a classic on creativity and productivity in research.
1 Early life and education
1.1 Childhood and family background
Richard Hamming was born in Chicago, Illinois, on February 11, 1915. His father, Charles Hamming, worked as a clerk, and his mother, Mabel Hamming, was a homemaker. He grew up in a modest household and showed an early aptitude for mathematics and science. As a child, he enjoyed solving puzzles and building models, traits that foreshadowed his later systematic approach to problem solving. He attended local public schools in Chicago.
1.2 Undergraduate studies at the University of Chicago
Hamming entered the University of Chicago in 1933, where he pursued a degree in mathematics. He was particularly influenced by the rigorous intellectual environment and the emphasis on foundational logic. Hamming earned his Bachelor of Science degree in 1937. His undergraduate work included exposure to the emerging field of mathematical logic, which later informed his approach to coding theory.
1.3 Graduate studies at the University of Illinois
Hamming continued his education at the University of Illinois Urbana-Champaign, where he earned a Master of Science degree in mathematics in 1939 and a Ph.D. in 1942 under the supervision of Waldemar Trjitzinsky. His doctoral dissertation was on linear differential equations with constant coefficients, a topic in numerical analysis. During his graduate studies, he also became interested in the use of calculating machines and mechanical methods for solving mathematical problems.
2 Career
2.1 Los Alamos National Laboratory (1945)
After completing his Ph.D., Hamming taught briefly at the University of Illinois. In 1945, he joined the Manhattan Project at the Los Alamos National Laboratory in New Mexico. There he worked on calculations for the atomic bomb, using early electromechanical and electronic computers. His experience with the unreliability of these early machines—they frequently made arithmetic errors—sparked his lifelong interest in error detection and correction.
2.2 Bell Labs (1946–1976)
2.2.1 Early work in numerical analysis and computing
In 1946, Hamming joined Bell Telephone Laboratories in Murray Hill, New Jersey. Initially, he worked on numerical analysis, developing improved methods for solving differential equations and for computing functions. He collaborated with other pioneers such as Claude Shannon and John Tukey. Hamming became known for his straightforward, practical approach to numerical computation, producing algorithms that were both efficient and robust. He also contributed to the design of the relay-based Bell Model V and later the IBM 604 and 650 computers.
2.2.2 Later role as head of the computing science department
By the late 1950s, Hamming had shifted his focus to error-correcting codes and information theory. In the 1960s, he became head of the Computing Science Department at Bell Labs, a position he held until his retirement from the Labs in 1976. Under his leadership, the department explored operating systems, programming languages, and numerical methods. Hamming was known for encouraging independent thinking and for his famous "no Monday morning quarterbacking" policy, which allowed researchers to take risks.
2.3 Teaching at the Naval Postgraduate School (1976–1997)
After retiring from Bell Labs, Hamming moved to Monterey, California, where he joined the faculty of the Naval Postgraduate School as a professor of computer science. He taught courses on numerical methods, coding theory, and the philosophy of research. He remained active there for more than twenty years, mentoring a new generation of engineers and scientists. His lecture series "The Art of Doing Science and Engineering" became especially influential.
3 Research contributions
3.1 Error‑correcting codes
3.1.1 Hamming codes
Working at Bell Labs in the late 1940s, Hamming became frustrated with the frequent errors made by the computers he used. He devised a systematic method to detect and correct single-bit errors using parity bits. The resulting family of codes, known as Hamming codes, are among the earliest error-correcting codes. The most famous example is the (7,4) Hamming code, which adds three parity bits to four data bits, allowing the correction of any single error. These codes are still used in memory (ECC RAM) and telecommunications.
3.1.2 Hamming distance
Hamming defined the Hamming distance between two binary strings of equal length as the number of positions at which the corresponding symbols differ. This metric is fundamental to coding theory and underlies the ability of error-correcting codes to distinguish valid codewords. The concept has applications beyond coding, including in cryptography, genetics, and machine learning.
3.1.3 Hamming bound and perfect codes
Hamming also established the Hamming bound (or sphere-packing bound), which gives an upper limit on the number of codewords that can exist in a binary code with given length and error-correction capability. Codes that achieve this bound are called perfect codes. Hamming codes themselves are perfect codes for single-error correction, and the only other nontrivial perfect binary codes are the Golay codes.
3.2 Numerical methods and scientific computing
3.2.1 Hamming window (Hamming–Tukey window)
Hamming, together with John Tukey, introduced a windowing function used in digital signal processing to reduce spectral leakage when performing the discrete Fourier transform. The Hamming window has the form w(n) = 0.54 – 0.46 cos(2πn/N) and is widely used in audio and speech processing, radar, and data analysis.
3.2.2 Numerical integration and differential equation solvers
Hamming developed several numerical methods for solving ordinary differential equations, including the Hamming predictor-corrector method. His 1962 book *Numerical Methods for Scientists and Engineers* became a standard reference, emphasizing practical, stable algorithms. He also contributed to the smoothing of numerical data and to the development of robust quadrature formulas.
3.3 Digital signal processing
Beyond the Hamming window, Hamming's work on digital filtering and spectral estimation influenced the field of digital signal processing. He advocated for a clear understanding of the relationship between continuous and discrete systems, and his emphasis on the use of computers for real-time filtering helped shape early signal-processing hardware.
3.4 Philosophy of research and creativity
Hamming is also remembered for his thoughtful reflections on what makes a successful researcher. His talk "You and Your Research" (given in 1986) and his book *The Art of Doing Science and Engineering* (1997) distill decades of observations. He stressed the importance of working on significant problems, maintaining courage, and having a "tolerance for ambiguity." His advice—such as "If you don't work on important problems, it's not likely that you'll do important work"—has inspired many in academia and industry.
4 Publications
4.1 Books
4.1.1 *Numerical Methods for Scientists and Engineers* (1962)
First published in 1962 and later revised, this textbook provides a thorough treatment of numerical analysis for engineering and science students. Hamming emphasized "numerical stability," "algorithmic efficiency," and the careful handling of errors. The book remains in print and is considered a classic in the field.
4.1.2 *The Art of Doing Science and Engineering* (1997)
Based on a course he taught at the Naval Postgraduate School, this book offers lessons on problem solving, creativity, and the practice of research. Hamming discusses the characteristics of great scientists, the role of luck, and the need for effective communication. It is often cited as an inspiring guide for young researchers.
4.2 Selected papers
Hamming published numerous influential papers, including "Error Detecting and Error Correcting Codes" (Bell System Technical Journal, 1950), which introduced Hamming codes. Other notable works include "Numerical Integration of Differential Equations" (with J. H. Wilkinson) and "On the Distribution of Numbers" (a study in computer arithmetic).
5 Awards and honors
5.1 IEEE Richard W. Hamming Medal (1988)
In 1988, the IEEE established the Richard W. Hamming Medal in his honor, awarded annually for exceptional contributions to information sciences and systems. Hamming himself was the first recipient of the medal, recognizing his seminal work in coding theory and digital signal processing.
5.2 Turing Award (1968)
Hamming received the ACM Turing Award in 1968 for his work in numerical methods, automatic coding systems, and error-correcting codes. The Turing Award is the highest honor in computer science, and Hamming's citation specifically mentioned his contributions to the Manhattan Project and his "pioneering efforts" in reliable computing.
5.3 Other recognitions
He was a Fellow of the IEEE and the ACM, and a member of the National Academy of Engineering. He received the IEEE Emanuel R. Piore Award in 1979 and was awarded an honorary doctorate from the University of Illinois. The Hamming Medal itself now bears his name.
6 Legacy
6.1 Concepts named after Hamming
6.1.1 Hamming code
Hamming codes are one of the most widely used families of error-correcting codes. They appear in computer memory (ECC), satellite communications, and RAID storage systems. Their simplicity and efficiency make them a standard textbook example.
6.1.2 Hamming distance
The Hamming distance is a fundamental concept in information theory, coding theory, and discrete mathematics. It is used in error detection and correction, as well as in fields such as bioinformatics (to compare genetic sequences) and machine learning (for nearest-neighbor classifiers).
6.1.3 Hamming window
The Hamming window is a standard tool in digital signal processing. It is employed in audio processing, speech recognition, and spectral analysis. The name "Hamming window" appears in virtually every introductory textbook on DSP.
6.2 Influence on computer science and engineering
Hamming's philosophy of research, codified in his talks and writings, has shaped the culture of computing research. His emphasis on tackling important problems, his willingness to question authority, and his practical approach to mathematics continue to inspire generations of scientists and engineers. Many of the numerical methods he developed are still used in computational science.
7 Personal life
7.1 Marriage and family
Hamming married Wanda Little in 1942. The couple had no children. Wanda supported him throughout his career, and they lived together in New Jersey during his Bell Labs years and later in California.
7.2 Hobbies and personal philosophy
Hamming was an avid woodworker and enjoyed building furniture. He also liked to take long walks, during which he would think about problems. He maintained a pragmatic and often contrarian outlook on life, frequently reminding his peers that "the purpose of computing is insight, not numbers." His personal philosophy emphasized curiosity, persistence, and a willingness to embrace failure as a learning opportunity.