The question depends on a definition
Asking who invented the computer is a bit like asking who invented the car. The answer depends on what you count. Do you mean the first mechanical design, the first working machine, the first electronic one, the first that could be reprogrammed freely, or the first that kept its program in memory? Each definition points to different people, and every one of them is reasonable.
It also helps to remember that “computer” originally meant a person who performed calculations. The term was later applied to machines that automated that work. So the history spans a long period during which the meaning of the word itself was changing.
A design before a machine
In the nineteenth century, Charles Babbage designed machines to automate mathematical tables. His proposed Analytical Engine, described in the 1830s and 1840s, was a mechanical design with ideas that look familiar today: a store for numbers, a unit that did arithmetic, and instructions supplied on punched cards. It was never completed in his lifetime, so it counts as a design, not a working computer. Ada Lovelace wrote extensive notes on it, which are often cited in discussions of early programming, though how much she contributed beyond explaining and extending the ideas is something historians discuss with care.
If your definition is “the first person to conceive of a general-purpose programmable machine,” Babbage is a strong candidate. If your definition requires something that actually ran, he is not.
Working machines in the 1930s and 1940s
The decades around the Second World War saw several independent efforts, often in different countries and sometimes in secret.
- Konrad Zuse, working in Germany, built a series of programmable calculating machines in the late 1930s and early 1940s. One of them is widely regarded as an early working programmable computer, using relays.
- John Atanasoff and Clifford Berry, in the United States, built an electronic machine for solving systems of equations. It was designed for a specific purpose rather than for general programming.
- Colossus, built in Britain during the war to help with codebreaking, used electronics at scale. It was programmable only in limited ways and was kept secret for decades afterward, so its role became known late.
- Harvard Mark I and similar electromechanical machines were developed in the United States, partly with industry support.
- ENIAC, completed in the mid-1940s in the United States, was a large electronic machine used first for military calculations. Its story is told in how the ENIAC paved the way for modern computing.
Which of these is “first” depends entirely on the adjectives. The earliest programmable one, the earliest electronic one and the earliest general-purpose electronic one are not necessarily the same machine. Disputes over credit even reached a US court in the 1970s, in a patent case that concluded certain ENIAC patent claims were not valid, in part because of earlier work. The ruling is itself sometimes argued over, and it is best seen as one legal view, not a final historical verdict.
The stored-program step
A separate and important milestone is the stored-program concept, in which instructions are kept in the same memory as data, so a machine can be reprogrammed by changing memory contents instead of rewiring or setting switches. The idea was circulating in the mid-1940s. A well-known written description came from a team connected with ENIAC’s successor, with the mathematician John von Neumann as the named author of a famous report, and that is why people sometimes talk about a “von Neumann architecture.” Others on the team also contributed, and credit is shared, a point that illustrates the problem with attaching one name.
Working stored-program machines followed in the late 1940s in Britain, with prototypes at Manchester and EDSAC at Cambridge among the early examples. Britain and the United States both produced early stored-program computers within a few years of each other.
If you define a computer as a machine that runs a stored program, the answer is a group of people in the late 1940s. If you stress that Alan Turing described a theoretical universal machine in 1936, you might credit mathematics as much as engineering; his paper was a concept, not a plan for hardware.
A table of definitions
| If “computer” means… | People or projects often mentioned |
|---|---|
| A general-purpose design | Babbage and Lovelace |
| An early working programmable machine | Zuse and others |
| A large-scale electronic machine | Colossus, ENIAC |
| A stored-program machine | Teams in Britain and the United States |
| A commercial product | Several firms in the late 1940s and 1950s |
The table is a simplification, and entries overlap, but it shows why no single row wins.
Why one name keeps getting chosen
Popular accounts favor a single figure because it is easier to remember. National pride, company histories and wartime secrecy also played roles: some work was classified, so its significance emerged late. And patents and publicity encouraged claims of priority. The more accurate picture is a web of overlapping efforts, with ideas moving between people, sometimes by contact and sometimes by independent invention. That pattern is explored further under collaborative invention.
A sensible answer to the title question is therefore a short list with caveats. The idea is traced to Babbage; early working machines to Zuse and several wartime projects; large electronic computing to Colossus and ENIAC; and the modern stored-program form to a group of teams in the 1940s. Anyone who gives you one name with confidence is probably relying on a particular definition. For broader context, the first computers and the electronic computers articles take up the machines in more detail.