A Field Before It Had a Name
Computer science did not begin as a subject of its own. Its roots lie in mathematics, logic and electrical engineering, and in practical problems that arose when people started building automatic calculating machines. For a while, people who worked with computers came from other departments and described themselves as mathematicians, physicists or engineers.
The questions they asked were nevertheless distinctive. What does it mean for something to be computable? How can a procedure be described exactly enough for a machine to follow it? How much time and memory does a task need? Over several decades, these questions grew into a recognizable discipline.
Theory in the 1930s
The theoretical foundations were laid in the 1930s, largely by logicians working on questions about the limits of mathematical proof. Several people, including Alan Turing and Alonzo Church, independently proposed precise definitions of what it means to compute something. Their formulations looked different but turned out to describe the same class of problems, which gave researchers confidence that the idea was well founded. Kurt Gödel’s earlier results on the limits of formal systems were part of this background.
An important result of this period was that some well-defined problems cannot be solved by any general procedure. This showed that computation has fundamental limits, and it remains a central theme of the field. In the late 1930s, Claude Shannon also showed how the logic of switching circuits, the ancestors of today’s logic gates, could be analyzed with Boolean algebra, connecting abstract logic to the design of real hardware. His later work in the late 1940s helped found information theory.
Machines Create Demand, 1940s and 1950s
Once electronic machines arrived in the 1940s, theory and practice began to meet. The people running early installations had to solve problems of numerical accuracy, memory organization and instruction design. Professional societies appeared; the Association for Computing Machinery was founded in the late 1940s, giving practitioners a place to share work. Journals and conferences followed during the 1950s.
In this period, much of the intellectual energy went into making machines usable. Researchers developed assembly languages, then higher-level programming languages and the translators that supported them. Work on computer algorithms, including methods for sorting and searching, began to be treated as a subject in itself. Early artificial intelligence research also emerged in the mid-1950s, bringing questions about reasoning and learning into the same community.
Becoming an Academic Discipline
Universities came to the subject gradually. In the 1950s, some institutions set up computing laboratories, often as service centers for researchers in other fields. Courses in programming and numerical methods were taught within mathematics, engineering or business programs. The University of Cambridge is often mentioned for offering a postgraduate diploma in the 1950s, and a number of universities in the United States and elsewhere followed.
During the 1960s, the idea of a separate department became common. Purdue and Stanford are frequently cited among the early universities to create computer science departments in the first half of the decade, though the exact history varies by country and institution, and other universities also have claims. The term “computer science” itself spread during these years, and it was not universally welcomed; some argued the subject was really engineering or mathematics, and others objected to the word “science.”
In the late 1960s, the ACM published recommendations for an undergraduate curriculum, which helped shape what students were expected to learn: programming, data structures, machine organization, and mathematical foundations. The ACM also began awarding its Turing Award in the mid-1960s to recognize major contributions to the field, a sign that the community regarded itself as a discipline with its own heroes.
Branching Into Subfields
By the 1960s and 1970s, the field had begun to divide into specialties. Some of the main ones were:
- Theory of computation and complexity, studying what can be computed and how efficiently.
- Programming languages and compilers, including formal ways of describing a language.
- Operating systems, which manage hardware resources and let several programs share a machine.
- Databases, including the relational databases introduced in research papers around 1970 and widely adopted later.
- Artificial intelligence, with periods of strong optimism followed by disappointment.
- Networking, as research on computer networking gave rise to the early internet.
Many of the field’s debates were about its identity. Was it a branch of mathematics, a type of engineering, or something new? In practice, it has drawn on all three.
Education and Wider Reach
The growth of computer science education beyond universities took time. Schools introduced programming to students from the 1960s and 1970s in some places, and the personal computer movement of the late 1970s and 1980s brought programming into homes. Over the following decades, enrollment in the subject grew and fell with the job market and with public interest, and curricula kept changing as the technology changed.
What the History Shows
The path from 1930s logic to modern departments shows a field defined as much by its questions as by its machines. Theory gave it ideas about limits and structure, and engineering gave it problems worth solving. Societies, journals and degree programs then gave those ideas an institutional home.
As with other histories in this area, details differ between countries, and the dates given here are approximate. What stays consistent is the pattern: a few theoretical insights, a wave of practical machines, and then universities building a shared curriculum around both.