Alan Perlis

Alan J. Perlis
Perlis in 1982
Born	Alan Jay Perlis (1922-04-01)April 1, 1922 Pittsburgh, Pennsylvania, U.S.
Died	February 7, 1990(1990-02-07) (aged 67) New Haven, Connecticut, U.S.
Alma mater	Carnegie Institute of Technology (BS) Massachusetts Institute of Technology (MS, PhD)
Known for	Internal Translator ALGOL Programming-language design Epigrams on Programming First Turing Award recipient
Awards	A. M. Turing Award (1966) IEEE Computer Pioneer Award (1986)
Scientific career
Fields	Computer science Programming languages Compilers
Institutions	Purdue University Carnegie Institute of Technology Yale University
Thesis	On Integral Equations, Their Solution by Iteration and Analytic Continuation  (1950)
Doctoral advisor	Philip Franklin
Doctoral students	David Parnas Zohar Manna Gary Lindstrom John R. Levine

Alan Jay Perlis (April 1, 1922 – February 7, 1990) was an American computer scientist, mathematician, and educator who helped establish computer science as an academic discipline. He was a pioneer in compiler construction and programming language design, and in 1966 became the first recipient of the A. M. Turing Award. The Association for Computing Machinery cited him "for his influence in the area of advanced programming techniques and compiler construction".¹

Perlis worked on early digital-computing projects after the Second World War, including Project Whirlwind at MIT and computing work at the Ballistic Research Laboratory. At Purdue University and the Carnegie Institute of Technology, he helped develop the Internal Translator (IT), an early algebraic compiler for machines such as the Datatron 205 and IBM 650.¹² He was also one of the American participants in the design of ALGOL 58 and later contributed to the growth of ALGOL-related language research.¹³

Beyond his technical work, Perlis was a major institution builder. He was the first editor-in-chief of Communications of the ACM, served as president of the ACM from 1962 to 1964, and helped shape early computer-science curricula.¹² At Carnegie, he became the first head of the graduate Department of Computer Science, one of the first such departments in the United States.¹³ He later moved to Yale University, where he held the Eugene Higgins Chair of Computer Science until his death.¹

Perlis was born on April 1, 1922, in Pittsburgh, Pennsylvania, into a Jewish family. He attended Colfax Public School in the Squirrel Hill neighborhood and entered Taylor Allderdice High School in 1933.¹ He studied chemistry at the Carnegie Institute of Technology, graduating with honors in 1942.¹

Two days after graduation, Perlis entered wartime service through the Aviation Cadet Meteorology Program of the United States Army Air Forces. He trained as a meteorology officer and later served in Europe as an intelligence and weather officer with a reconnaissance squadron.¹ After the war he briefly pursued graduate study in chemistry at the California Institute of Technology, but soon shifted to mathematics at the Massachusetts Institute of Technology.¹

At MIT, Perlis studied under mathematician Philip Franklin. He earned a master's degree in mathematics in 1949 and a doctorate in 1950. His dissertation was titled On Integral Equations, Their Solution by Iteration and Analytic Continuation.¹

Perlis's entry into computing came through MIT's Project Whirlwind, one of the early real-time digital-computer projects. Franklin's mathematics group was responsible for preparing programs for Whirlwind, and Perlis assisted during the summers of 1948 and 1949 with coding and with evaluating numerical methods suitable for the machine.¹ In 1951 he worked at the Ballistic Research Laboratory at Aberdeen Proving Ground, where he computed ballistic tables and encountered one of the early multi-machine computing environments, including ENIAC, EDVAC, and ORDVAC.¹

In 1952, Perlis returned to MIT's Digital Computer Laboratory and worked on Project Cape Cod, an experimental air-defense system that helped lead to SAGE.¹ Later that year he joined Purdue University as an assistant professor of mathematics and director of the computational division of the university's Statistical Laboratory.¹²

At Purdue, Perlis became interested in making programming less dependent on direct machine coding. He helped lead the development of a "mathematical language compiler" for the Datatron 205, later known as Internal Translator or IT.¹ Unlike many early programming systems, IT was designed with a logical structure that could be moved to other machines with relatively few changes, making it an early step toward machine-independent programming languages.¹

In 1956, Perlis moved to the Carnegie Institute of Technology as associate professor of mathematics and director of the computation center. There he and collaborators completed an implementation of IT on the IBM 650.¹² The IEEE Computer Society describes Perlis and his colleagues as having built a sequence of algebraic-language compilers and assemblers for the IBM 650.² IT was adopted by a number of university computing installations and helped stimulate research into programming techniques and higher-level languages.¹

In the late 1950s, Perlis became active in efforts to define a machine-independent programming language. In 1957, user groups and the ACM began considering a "universal programming language" that could express numerical methods independently of particular computers.¹ In May 1958, Perlis led the American ACM delegation to a meeting in Zürich with a German-Swiss group of programming-language specialists. The meeting produced the International Algorithmic Language, later renamed ALGOL.¹³

ALGOL 60 became one of the most influential programming languages in computing history. It helped establish block structure, lexical scoping, and formal language description as central topics in programming-language design. Although Perlis was not the only designer, his work on IT, his role in the ACM delegation, and his later advocacy helped make programming languages a legitimate object of academic study.¹

At Carnegie, Perlis also helped build computer science as an institutional discipline. He became chair of the mathematics department, co-directed a graduate program in systems and communications science, and in 1965 became the first head of the graduate Department of Computer Science.² ACM credits him with helping to secure ARPA support that aided the launch of Carnegie's computer-science department.¹

Perlis was one of the early leaders of the Association for Computing Machinery. He served as the first editor-in-chief of Communications of the ACM from 1958 to 1962 and as ACM president from 1962 to 1964.¹² During his ACM presidency, the organization established a curriculum committee on computer science. The committee's work contributed to the first recommendations for undergraduate computer-science programs.¹

In 1966, Perlis became the first recipient of the A. M. Turing Award. His award citation recognized his influence on advanced programming techniques and compiler construction.¹ The award connected his early compiler work, his role in ALGOL, and his advocacy of programming as a central subject in computing.

In 1971, Perlis moved to Yale University as Eugene Higgins Professor of Computer Science. He remained at Yale until his death, except for the 1977–1978 academic year, when he was Gordon and Betty Moore Professor of Computer Science at the California Institute of Technology.¹³

At Yale, Perlis continued to write and teach about programming languages, computing education, and the nature of programming. His later publications included work on APL, software metrics, and functional approaches to array-oriented programming.² He also supervised or influenced students who became important figures in programming languages and software engineering, including David Parnas and Zohar Manna.

Perlis is also remembered for "Epigrams on Programming", a collection of short aphorisms first published in ACM SIGPLAN Notices in 1982.⁴ The epigrams compressed Perlis's views on programming, language design, abstraction, and software culture into short, often humorous statements. They were widely circulated among programmers and computer scientists.³

One of the best-known epigrams introduced the phrase "Turing tarpit", referring to a language or formal system in which "everything is possible but nothing of interest is easy".⁴ The phrase became a common criticism of programming systems that are computationally universal but practically awkward.

Perlis received the inaugural A. M. Turing Award in 1966.¹ He was elected to the American Academy of Arts and Sciences in 1974 and to the National Academy of Engineering in 1977.¹ His additional honors included the AFIPS Education Award in 1984 and the IEEE Computer Pioneer Award in 1986.¹² He received honorary doctorates from Davis and Elkins College, Purdue University, the University of Waterloo, and Sacred Heart University.¹

Perlis died in New Haven, Connecticut, on February 7, 1990.¹ In a 1990 memorial article in Communications of the ACM, Peter J. Denning described him as "a founding father of computer science as a separate discipline".⁵

Perlis's legacy lies in both technical and institutional work. Technically, he helped move programming from machine-specific coding toward higher-level languages and compiler systems. Institutionally, he helped create the organizations, curricula, journals, and departments through which computer science became a recognized academic field.¹²³

• Perlis, Alan J.; Smith, J. W.; Van Zoeren, H. R. (1957). Internal Translator: A Compiler for the 650.
• Perlis, A. J.; Thornton, C. (1960). "Symbol Manipulation by Threaded Lists". Communications of the ACM. 3 (4): 195–204. doi:10.1145/367177.367202.
• Perlis, Alan J.; Braden, Robert T. (1965). An Introductory Course in Computer Programming.
• Galler, Bernard A.; Perlis, Alan J. (1970). A View of Programming Languages.
• Perlis, Alan J. (1975). Introduction to Computer Science.
• Perlis, Alan J. (1977). In Praise of APL: A Language for Lyrical Programming (Report).
• Perlis, Alan J. (1978). Almost Perfect Artifacts Improve Only in Small Ways: APL Is More French than English (Report).
• Perlis, Alan J.; Sayward, Frederick; Shaw, Mary (1981). Software Metrics: An Analysis and Evaluation.
• Perlis, Alan J. (1982). "Epigrams on Programming". ACM SIGPLAN Notices. 17 (9): 7–13.
• Perlis, Alan J.; Tu, Hai-Chen (1986). FAC: A Functional APL Language. APL '86 Conference Proceedings.