A machine designed, but never built
In the 1830s, an English mathematician and inventor named Charles Babbage began sketching a machine that was meant to do far more than any calculator of its day. He called it the Analytical Engine. It was to be a mechanical, steam-powered device that could carry out long sequences of arithmetic under the direction of a program, and then repeat or vary that sequence as needed. No complete Analytical Engine was ever built in his lifetime, and he kept revising the design for decades. Even so, its plan contains several ideas that readers will recognize in later machines.
It is worth being careful with the word “first.” The Analytical Engine was a design, not a working computer, and the question of what counts as the first computer depends heavily on the definition used. The Analytical Engine is best described as an early and remarkably complete blueprint for a general-purpose, programmable machine.
From the Difference Engine to something more general
Babbage’s earlier project, the Difference Engine, was a special-purpose machine meant to calculate and print mathematical tables, using a method that needed only repeated addition. Accurate tables mattered at the time for navigation, astronomy and engineering, and hand-calculated ones often contained errors. Work on the Difference Engine received government support in the 1820s and 1830s but stalled, partly because of the difficulty of building such precise parts and partly because of disputes over cost and control.
While that work was stuck, Babbage realized that a machine could be built to follow instructions rather than a single fixed procedure. That shift, from a machine for one job to a machine for any calculable job, is the heart of the Analytical Engine.
The store and the mill
The design divided the machine into parts that map closely onto later ideas. Babbage called them the store and the mill.
- The store was where numbers were held. It was intended to hold a large number of values, each as a many-digit decimal number represented by a column of rotating wheels. It served the role that we now call main memory.
- The mill was where arithmetic took place. Numbers were brought from the store to the mill, operated on, and the results sent back. In modern terms the mill resembles the arithmetic part of a central processing unit.
The names came from the textile industry, where a mill processed material that had been kept in a store. Babbage also planned output devices, including a printer and a device that could make a stamped record, so that results would not need to be copied by hand and risk new errors.
One difference from modern machines is worth noting. Babbage worked in decimal, with wheels marked from zero to nine, rather than in binary numbers, which later electronic machines largely adopted because two-state electronic parts were easy to build and reliable.
Control by punched cards
The control system was borrowed from weaving. In the early 1800s, Joseph Marie Jacquard had introduced a loom whose patterns were selected by chains of punched cards. Babbage proposed using similar cards for the Analytical Engine, in two roles: some cards would specify the operations to be performed, and others would specify which numbers in the store were involved. Because the cards could be arranged in different sequences, the same machine could in principle be given different programs without changing its mechanism.
The design also described conditional behavior, where the machine could change what it did next depending on a result, and repetition of a group of steps. These features are what separate a programmable machine from a calculator that simply follows one fixed path. The cards also hold the program separately from the numbers in the store, which differs from the later stored-program concept in which instructions and data share the same memory. That later idea was developed in the 1940s and is not part of Babbage’s plan.
Lovelace’s notes
The best-known early description of the machine’s possibilities came from someone other than its inventor. In the early 1840s, the mathematician Luigi Menabrea published a French-language account of a lecture Babbage had given in Turin. Ada Lovelace translated it into English in 1843 and added her own notes, which were considerably longer than the original text.
Those notes are often cited for two reasons. One is a worked table of operations for calculating a sequence of numbers known as Bernoulli numbers, often described as an early published program for a machine. The extent of her personal contribution to the details of that table, compared with Babbage’s, has been debated by historians, and the safest summary is that the two worked closely together. The other reason is her observation that such an engine might manipulate things other than numbers, such as musical notes, if their relationships could be expressed in the machine’s terms. She also cautioned that the engine had no pretensions to originate anything and could only do what it was told to do.
Why it was never completed
Several factors combined, and historians generally point to a mix rather than a single cause.
- Precision manufacturing. The machine called for thousands of parts made to tolerances that were difficult to achieve with the tools of the time.
- Funding and politics. After the frustrations of the Difference Engine, the British government was unwilling to commit more money, and private support was limited.
- Disputes and shifting designs. Babbage was often in conflict with his engineers and officials, and he kept improving the plans rather than freezing one design to build.
- Scale. The planned machine was very large, and a complete version would have been an expensive and complex undertaking.
It is sometimes claimed that the design was flawed and could never have worked. The evidence does not support so strong a statement, and it is more cautious to say that it was never tested in full. In 1991, the Science Museum in London completed a working Difference Engine No. 2 built to Babbage’s later drawings, which suggests that his mechanical thinking was sound for that machine.
What the Analytical Engine did and did not influence
There is little evidence that the builders of the electronic machines of the 1940s drew directly on Babbage’s work, and several groups arrived at similar ideas independently. His reputation was largely revived afterward, when historians and computer pioneers looked back at his papers. Alan Turing’s later theoretical work on general-purpose computation, for example, developed along its own path, although the parallels are striking. It is fair to say that Babbage anticipated the structure of a general-purpose computer, while the people who built the first computers worked out their solutions in a different era and with different technology.
The Analytical Engine remains important as a demonstration that the basic logic of programmable computing, a memory, a processor, an input and output system and a stored sequence of instructions, could be imagined well before any technology existed to make it practical. Whether that makes Babbage the originator of the computer is a matter of definition, a question taken up in the discussion of who invented the computer.