Undecided about whether to pursue art or engineering, he pursued engineering, but with a continued interest in design—for his senior school project, he had designed a city of the future (à la Fritz Lang’s Metropolis) based on a hexagonal grid. Like Alan Turing, Zuse was educated in a system that focused on a child’s emotional and philosophical life as well as his intellectual life, and at the end of school, like Turing, Zuse found himself to be something of an outsider—to the disappointment of his very conventional parents, he no longer believed in God or religion.
In 1927, at the same time that Turing was making his difficult way through the Sherborne School, Zuse entered the Technische Universität in Berlin and took up residence in the city of his birth, a sociable young man in an exciting and rapidly changing urban environment. He was immediately fascinated once again by the bridge building that was going on, a fascination that was encouraged by the requirement that students at the Technische Universität had to have practical experience in ironwork or carpentry or bricklaying. Zuse’s experience in these trades served to break down class barriers somewhat, but he remained a thinker more than a builder—interested in photography, movies, drawing, performance. When he became intrigued by a technological question, such as how to build a rocket that might head to a distant star, it was more often through some form of art, such as science fiction, than it was through science itself. He does not mention taking an interest, per se, in physics or mathematics or cosmology, as Atanasoff and Turing did. He writes, “Given my many detours and by-ways, I am still amazed that I earned a diploma at all.” (And then he goes on to recount how he was lucky that his mathematics examiner asked a particular question—as he was eavesdropping upon the questions that the other students were asked, he realized that he could not have answered any of them.)
According to Zuse, amid all the busyness, freedom, and pleasure of his university and postuniversity life, there was not much understanding about what the Nazis were up to. While Zuse himself was reading Das Kapital and the autobiography of Henry Ford, neither he nor his friends paid much attention to those who were reading Mein Kampf. Zuse, as a son of the Prussian civil servant class, felt more inspired by the writings of Oswald Spengler than of Hitler, especially Spengler’s anti-Marxist 1920 political tract Prussianism and Socialism. Even so, Zuse found the Marxists he knew friendly and interested in discussion.
Times soon changed, and “on all sides now Germans were being forced into line and marched off,” yet Zuse and his fellow students seem to still have had the feeling that they had some freedom of opinion, some future in terms of working choices. And then, on the night of June 30 (the Night of the Long Knives), Hitler used his personal bodyguard, the SS, to purge Ernst Röhm, an enemy in the von Hindenburg government, and two hundred of his allies in the armed forces. When von Hindenburg died a month later, Hitler made himself president and head of the armed forces, which were henceforth to pledge allegiance to him personally rather than simply to the state. After von Hindenburg’s funeral, Hitler assumed the title “führer,” but, perhaps as an indication of Zuse’s ongoing focus on other things, he writes, “The psychological effect was that one assumed the impetuous and hysterical period would now be followed by a period of common sense and work.” Zuse belonged to a fraternity of long standing at the university. When the three Jewish members were required by the Hitler government to leave the club, the club decided to disband but ultimately did not do so, only because the Jewish members asked them not to.
At the end of his university career, Zuse idled about Berlin for a year, undecided about what to do next, but in 1935, aged twenty-five, he took an engineering job with the new aircraft division of Henschel and Son, a locomotive corporation that was to produce several types of planes for the Luftwaffe. Zuse, apparently alerted by his new job to the sorts of calculating problems that aircraft design required, quit almost immediately to begin his own project—a computer.
According to Zuse’s account, he started from scratch. “When I began to build my own computer, I neither understood anything about computing machines nor had I ever heard of Babbage.” Zuse is not clear about why he decided to build a computer, or the theoretical basis of the machine, but it seems to have grown out of his talent for and interest in design rather than the desire to solve a particular kind of mathematical problem. His first attempt at a machine had nothing to do with mathematics—it was a skeletal vending machine “which took money and gave mandarin [oranges], and sometimes, indeed, returned the money with the mandarins.”
Zuse’s working space was the living room of his parents’ apartment and his capital (amounting to several thousand marks at the most) came from his father’s and sister’s paychecks and the contributions of his friends who had managed to find jobs or had a bit of money. His collaborators were his friends from the technical school, who received their pay in the form of meals that Zuse’s mother provided. His raw materials were bought piecemeal when he had the money, and they were simple ones. One friend describes how he made the mechanical relays—Zuse would draw the pattern on a piece of paper, then the friend “pasted the paper on a small plywood board, then fixed the necessary number of metal sheets between it and a second board that lay under it.” He then “screwed the two boards together with threaded screws, and sawed out the form of the relays with a small, electric fretsaw.” He “made these relays by the thousand.”1
Zuse seems to have built on the loyalty he developed in his college fraternity to accrue dedicated student helpers, who, like Atanasoff and his students (and Turing, too) recognized that the sorts of calculations they were required to do normally with analog desk calculators were much easier with Zuse’s machine. But the project was secret (Zuse does not say why, but possibly the authorities would have looked with suspicion upon a project that was diverting parts and supplies from war preparations). Those working on it declared when asked that they were attempting to build an aircraft tank gauge, because the German Air Ministry was at the time sponsoring a contest to build such a machine.
The basis of Zuse’s original design was electromechanical, akin to telephone relays, with which Zuse happened to be familiar, but, like other pioneers, he soon realized that the number of relays required in even a small-capacity machine was impractically enormous—the machine he was building so filled the family apartment that a friend who was working on it later wrote, “It took up almost the entire living room. It was a permanent fixture in the apartment. I think that it was only after the house was bombed during the war that the first Zuse Universal Computing Machine could be moved into the museum.” In his first effort, Zuse had some success with his electromechanical ideas and was able to build a flexible enough device so that he could use it to test his ideas about switching and build his understanding of mathematical logic.
As his work progressed, Zuse decided he needed more reliable financing. In 1937, he got in touch with a Dr. Kurt Pannke, who manufactured calculators. Pannke told the young man, “I don’t want to discourage you from continuing work as an inventor and from developing new ideas, but I must go ahead and tell you one thing: in the field of computing machines, practically everything has been researched and perfected to the last detail.” When Zuse told Pannke that his prototype could multiply, Pannke was silent for a long time and then came for a visit to the machine. Zuse demonstrated that because (like Atanasoff) he was using binary numbers (only the digits 1 and 0), adding and multiplying amounted (literally) to the same thing. In his autobiography, Zuse demonstrates why this is. When a calculator uses ten digits (0–9), the number of different keys required to represent the multiplication table is unwieldy—0 × 1 = 0, 2 × 2 = 4, 6 × 6 = 36, 8 × 8 = 64, with each digit represented by a key of its own. As we saw in third grade, when we were learning the multiplication tables in the back of our arithmetic books, between 0 × 0 and 9 × 9, there are a hundred different numbers. In a binary system, 0 × 0 = 0, 1 × 0 = 0, 0 × 1 = 0, and 1 × 1 = 1. Only two digits are needed. The problem for Pannke, as a busi
nessman, was that calculators that multiplied by repeated adding were cheaper to build than calculators that attempted to multiply; there was a limited market for calculators, so adding was good enough. Zuse points out, “To construct large and expensive computing machines for scientists, for mathematicians and engineers, appeared absurd, and above all held no promise of commercial success. These people didn’t have any money.” But Pannke gave Zuse about 7,000 reichsmarks, and he began to work on his second prototype, the Z2.
In his home workshop and at school, as well as in plans and diagrams, Atanasoff was trying this and that. The work was taxing and frustrating mostly because there was no apparent place to begin. Every idea he came up with immediately branched into a tangle of relationships that were complex and contradictory. And he had to factor in available hardware. Like other inventors of the computer, he knew that rods and gears and motors were reliable and much more precise than they had ever been, thanks to advances in machining and production—Atanasoff was tempted by these advances to pursue the analog path. But he was strongly drawn to the speed that the novel, but as yet unreliable, technology of electronics offered.
Atanasoff’s interest in binary system was not based on quite the same reasoning as Zuse’s interest—IBM had, after all, introduced a multiplying calculator in 1931. What he suspected was that using a binary number system would make it possible to use vacuum tubes for actual calculating. The vacuum tubes would be arranged inside a processing unit and different arrangements of on and off tubes would stand for different numbers—any number could be represented by a row of on-off vacuum tubes. At the same time, although he himself was perfectly familiar with binary counting systems, he knew that not many other people were (something Zuse’s experience also demonstrated)—even most mathematicians were uncomfortable operating outside of the decimal system. The prevailing wisdom was that translating from the binary system to the decimal system would pose an enormous difficulty—a decimal number would have to be entered somehow, turning on the tubes and turning them off, then, when the calculation had been performed, the result would have be communicated to some sort of output mechanism that would translate the binary number to a decimal number.
And there would have to be “memory.” In the most advanced IBM tabulator of the day, there were two types of memory. The first comprised the set of instructions that the tabulator used to carry out operations. If a byte in today’s computer terminology consists of 8 bits of storage capacity,2 the first type of memory belonging to the IBM tabulator of Atanasoff’s day had 266 bits or 33.25 bytes of memory. That tabulator’s memory hardly bears comparison with what we are familiar with in 2010—a single page of text saved as a file in Microsoft Word that includes all preference settings, containing 514 words and forty lines of English text in 14-point type, uses 28 kilobytes, or 28,662 bytes, or 862 times the capacity of the IBM tabulator Atanasoff was familiar with. A single 3.2 megabyte digital photograph (3.2 million bytes) uses almost 96,400 times the IBM’s memory capacity. The IBM’s second type of memory was larger, but external to the machine—it was the record of calculations produced, punch cards that could then be fed back into the machine and used for future tabulations. The punch cards, of course, were kept track of by the operator, not the machine.
Modern computers still have two types of memory. The first type is called the RAM, or random-access memory, which the computer uses while it is turned on for operations, applications, and frequently accessed data. The second type is the storage memory, which the computer has access to and is stored externally to the main operating system on hard disk drives, floppy disks, magnetic tape, and so on. Although today, at least in personal computers, they are both inside the computer, the two kinds of memory follow Atanasoff’s (and IBM’s) ideas by being separate but communicating with each other.
When Atanasoff jumped in his new Ford V8 that evening in December 1937, he later testified, “I was in such a mental state that no resolution was possible. I was just unhappy to an extreme degree.” But he was pleased with his new car (Burton notes that he purchased a new car every year). He enjoyed its speed and maneuverability. He felt himself calm down, and he also felt a sort of suspension of time—“When I finally came to earth I was crossing the Mississippi River, 189 miles from my desk.” His next thought was perhaps characteristic of his practical and no-nonsense temperament: “Now you’ve got to quit this damned foolishness.”
Then he saw the tavern sign. He went in, sat down, and ordered a bourbon and soda. A radio sitting behind the bar was playing music. Almost as soon as the waitress brought him his drink, the nature of his computing system occurred to him as a logical whole, and he began envisioning both the component pieces and how the pieces could fit together. He jotted some notes down on a paper napkin, but later he didn’t need the notes because he was able to visualize and contemplate his machine so thoroughly that he had no trouble recalling what he had come up with. He sat in the bar for several hours, thinking through each of his concepts but concentrating particularly upon ideas for how the memory would work and how an electronically based on-off process would calculate.
Atanasoff’s experience is interesting on a number of levels. The way in which a state of effort followed by a state of relaxation induced an understanding of the system he wanted to build is reminiscent of what had happened to Turing and also to Henri Poincaré, the mathematician, as quoted in psychiatric researcher Nancy Andreasen’s The Creative Brain:
For fifteen days I strove to prove that there could not be any functions like those we have since called Fuchsian functions. I was very ignorant: every day, I seated myself at my work table, stayed an hour or two, tried a great number of combinations and reached no results. One evening, contrary to my custom, I drank black coffee and could not sleep. Ideas rose in crowds; I felt them collide until pairs interlocked, so to speak, making a stable combination. By the next morning, I had established the existence of a class of Fuchsian functions … I had only to write out the results, which took but a few hours.
But what Poincaré really wants to do is to boil his results down into a principle that can be understood in relation to other well-known mathematical principles. When he then takes a trip, he manages to do this without even interrupting his conversation with another passenger: “The changes of travel made me forget my mathematical work. Having reached Coutances, we entered an omnibus to go someplace or other. At the moment when I put my foot on the step, the idea came to me without anything in my former thoughts seeming to have paved the way … On my return to Caen, for conscience’s sake, I verified the result at my leisure.”
Andreasen then goes on to detail recent research (as of 2005) into how the brain is structured and how it works to create. She describes the brain as a system of sending and receiving neurons that are organized into areas that govern different functions. They connect to one another at synapses, where a tiny electric charge jumps over a tiny space. The neurons are embedded in gray matter (the cortex of the brain that contains nerve cell bodies), and the fuel of the brain is glucose. Andreasen distinguishes between ordinary creativity of the sort that is required in talking and the extraordinary creativity required for innovative or artistic thought. She points out that “most of the time that we speak, we are producing a sequence of words that we have not produced before.” But the sort of creativity that invents the computer is of a different order. The brain, she argues, is a self-organizing system “created from components that are in existence and that spontaneously reorganize themselves to create something new.” An essential part of a self-organizing system is the feedback loop—in the brain, this would consist of electrical impulses passing along neurons back and forth between one part of the brain and the others, contradicting or reinforcing earlier impulses and influencing later ones.
In order to understand how the brain creates, Andreasen distinguishes between episodic memory, used for personal reminiscence and free association of thoughts, and semantic memory, used for information storage and
retrieval of thoughts and concepts not related to personal history. Using positron emission tomography (PET) to image her study subjects’ brains while they relax and free-associate, Andreasen discovered that the most active regions in her free-associating subjects’ brains were the associative regions, that is, the frontal, parietal, and temporal lobes, the most complexly structured regions, the slowest to develop, and the regions dedicated to generating connections among all the other regions of the brain. She notes that in famous recollections of creative moments by poets such as Coleridge and scientists such as Poincaré and chemist Friedrich Kekulé (who dozed off and dreamed of a snake eating its tail and came up with the structure of the benzene ring), there is often a sudden flash of insight, in which previously unconnected ideas combine into a new thing. She explains this often attested experience: “I would hypothesize that during the creative process, the brain begins by disorganizing, making links between shadowy forms of objects or symbols or words or remembered experiences that have not previously been linked. Out of this disorganization, self-organization eventually re-emerges and takes over in the brain. The result is a completely new and original thing.”
Clearly, Atanasoff began his trip from Ames, Iowa, to Rock Island, Illinois, in a disorganized (and frustrated) state. Like Turing and Poincaré, though, once he was able to forget his mathematical work, ideas that had refused to come together when he was thinking about them (using his semantic memory) succeeded in coming together once he came to earth upon crossing the Mississippi and realized how far he had traveled in a dreamlike state.