As We May Think: Vannevar Bush, WW2, and Tech Today

Friday, May 31, 2019
Christopher Sirk

Today, we no longer think about the array of complex devices around us, underpinning almost every aspect of our lives. These invisible systems only get any attention when they go out of whack.  

We take it for granted our laptop will turn on, our Wi-Fi will work, our Amazon Prime delivery will arrive the next day. We sit on a budget flight whisking us halfway across earth and fiddle with work emails, while the guy next to you binge-watches every movie starring The Rock available.

The roots of the system building that connects all this ubiquitous tech can be traced to the dark days of World War II and the research culture that helped win the war for the Allies.

The story of American engineer and administrator Vannevar Bush, who led American science full stop for about seven years, is an important piece of this puzzle. And it is a story that keeps on giving, with fingerprints on everything from the creation of the internet to cloud-based apps.


Vannevar Bush and the rise of the machines

“Consider a future device …  in which an individual stores all his books, records, and communications, and which is mechanized so that it may be consulted with exceeding speed and flexibility. It is an enlarged intimate supplement to his memory.”

— Vannevar Bush

Born in Everett, Massachusetts in 1890, Vannevar Bush was primed for a life of overachievement. Following in his father’s footsteps, he attended the prestigious Tufts College, where he earned a Bachelor’s and Master’s of Science in four years, graduating in 1913 having invented and patented a piece of land surveying equipment called the “Profile Tracer” along the way. After a few years working at General Electric and teaching at Tufts, he went on to pursue a doctorate in electrical engineering from a joint MIT-Harvard program, which he finished in 1916.

Afterward, Bush got another job at Tufts and started working with the American Radio and Research Corporation (AMRAD). He left Tufts in 1919 to join the Department of Electrical Engineering at MIT (although he kept working with AMRAD).

From 1927 to 1931, Bush produced the world’s first analog computer at MIT, the Differential Analyzer. A direct ancestor to the modern computer, it was powered by electricity, yet highly mechanical, using drive shafts and gears to represent complex equations. The machine could solve differential equations with up to 18 variables.

It was clunky, enormous, and took ages to set up. But by speeding up calculations dramatically, the Differential Analyzer awakened scientists and engineers to the advantages that more powerful computers could bring.

Bush’s work spurred the building of Differential Analyzers elsewhere. English mathematician and physicist Douglas Hartree built one in England in 1935. Engineer Sasaki Tatsujiro and others built one in Tokyo in 1942, and in 1948 Beatrice Worsley (the first person ever awarded a PhD in ‘Computer Science’) built one in Toronto too.

Powerful, simple, and adaptable, the machine was used in the 1930s by General Electric’s Edith Clarke (the first female electrical engineer at the company) to analyze power lines and electrical infrastructure. During World War II it was used to calculate missile trajectories, and (probably most importantly) as a round-the-clock calculator for radiation research at MIT as part of the Manhattan Project.

The Differential Analyzer raised Bush’s profile considerably, gaining the ears of higher-ups in the military, academia, and government.

By the end of the 1930s, as storm clouds gathered over Europe and Japanese aggression in China continued unabated, Bush dreamed about putting together a centralized research organization to innovate better than hostile nations.

In June 1940, as the German army blitzed their way to Paris faster than anyone could’ve predicted, Bush gained an audience with President Franklin Delano Roosevelt and made his case for said organization. Roosevelt approved almost immediately, and the National Defense Research Committee (NDRC) was created, which quickly morphed into a subset of the Office of Scientific Research and Development (OSRD), headed by Bush.

Virtually overnight, Bush became the most powerful scientist in America. At its peak, OSRD directly employed 6,000 researchers working on a budget of $3 million a week (by the end of the war, they’d spent $450 million dollars or about $6 billion in today’s currency). The organization pooled talent and resources from the US military with that of universities like MIT, Harvard, and UC Berkeley, fostering a culture of information sharing and cooperation.

The OSRD oversaw a number of extremely important, war-tide turning and world-changing technological advancements. Radar technology (developed from initial research by cash-strapped British scientists, who gave their work to OSRD) helped win the naval war against German U-Boats, proximity fuses made Allied artillery more effective, and atomic research gave the Americans the most destructive weapon in the world, first.

Bush became a significant public figure during the war. In April 1944, he appeared on the cover of TIME. In July 1945, as the war wound down, he published an influential article in The Atlantic about the future of technology and the legacy of the OSRD’s research culture; it was titled “As We May Think.”

“As We May Think”

“A record if it is to be useful… must be continuously extended, it must be stored, and above all it must be consulted.”

— Vannevar Bush

In “As We May Think,” Bush unsurprisingly makes a strong case for the cause of scientific research, but more presciently, he argues that the dissemination of said research is what is truly important. He predicts that the future of scientific advancement is bound up in the development of new forms of information recording, retrieval, and sharing.

To that end, he structures his essay around the concept of a hypothetical mechanized private file archive and library, which he dubs a “memex.”

Bush’s memex is a system that mimics the associative processes of the human mind, yet unlike these processes is available as a permanent record. It builds trails through information for humans to navigate and leaves that trail available to review and consult later.

In his article, Bush writes that with the memex, an enormous amount of information, stored on photocells and microfilm, we will be able to review and access data linked by “associative trails” at a high rate of speed. This recall will also, he suggests, “beat the mind decisively in regard to the permanence and clarity of the items resurrected from storage.”

The creation of the memex is necessary for us to cope with the sheer wealth of information being produced. Bush’s essay speaks specifically about trying to keep track of all the scientific publications out there, which by the 1940s were already making it hard to make real, rigorous use of the total scientific record.

Our inability to access the most important information in the least amount of time ultimately slows the dissemination of ideas. Consequently, the refinement of ideas and their mainstream adoption takes longer than it should.

Bush gives several examples of how brilliant innovations, without an economy to back them up, are in the end not very useful. He brings up German mathematician and philosopher Gottfried Leibniz’s calculating machine, the (ever-so-excellently named) Step Reckoner. Designed in 1671 and built in 1673, the mechanical counting machine was, in essence, the first four-function calculator, using drum shaped gears that could rotate to represent digits from 0 to 9.

It was absolutely groundbreaking. So much so that it was completely impractical from the perspective of mainstream adoption.

The industrial revolution of the late 1700s was still many years off, and mass production simply did not exist. As such, the money, time, and labor required to manufacture and maintain the complex device exceeded the resources needed to employ a mathematician using quill and paper. Subsequently, the Step Reckoner failed to incite any kind of broader societal transformations.

Fast forward to the 1940s, the decade of Bush’s “As We May Think” publication, and complex devices like the typewriter, the movie camera, the automobile, and any number of other advanced mechanical products were all ubiquitous to everyday life, fairly cheaply available, and performed reliably. Whereas before modern technology and manufacturing, complexity had been synonymous with unreliability, now complexity could be layered on top of complexity, with the (fairly) safe assumption that the previous layer of technology has been mastered and made inexpensive. This set the stage for a major intensification of innovation, all predicated on “modification and extension.”

Bush was excited about new tech, speaking gushingly about photocell development allowing devices to see for us, and automated electronic relays that carry out sequences of movements or operations faster and more accurately than any human operator ever could.

He recognized the keyboard-operated Voder voice synthesizer, first presented to the public at the 1939 World’s Fair in New York, as the world’s first text-to-speech machine. In turn, he predicted that the opposite, a speech-to-text device, could be developed by suggesting a merging of IBM’s Vocoder technology and the stenotype speech transcription machine used in courtrooms.

He also uncannily predicted the invention of the Polaroid two years later when he described a hypothetical technology that produces an instant photographic record with a single click.

The net effect of instrumentalizing all this technology in the service of record keeping would be to allow humanity to access what he calls its “shady past,” and to “analyze more completely and objectively [its] present problems.”

At the same time, the building of such a great record, easily accessible to all, would free us to “reacquire the privilege of forgetting the manifold things [one] does not need to have immediately at hand, with some assurance that [one] can find them again if they prove important.”

From WW2 to WWW: Bush and Tech Today

“Much needs to occur… between the collection of data and observations, the extraction of parallel material from the existing record, and the final insertion of new material into the general body of the common record. For mature thought there is no mechanical substitute. But creative thought and essentially repetitive thought are very different things. For the latter there are, and may be, powerful mechanical aids.”

— Vannevar Bush

The German media theorist Friedrich Kittler linked the decisive development of cybernetics, the study of how information is communicated electronically and mechanically, with the computing machines developed in the USA under the aegis of Bush, as well as English scientist Alan Turing’s code-breaking work at Bletchley Park.

For Kittler, who famously stated that “media determine our situation,” cryptanalysis, code-breaking, and computers defeated Germany in World War II (probably a bit simplistic given the Soviet Union’s massive contribution). By winning the war, these technologies became pre-eminent, and in turn invaded civilian life, as military applications slowly but surely seeped into the realm of commercial products, as is their custom.  

In 1947, Bush's beloved OSRD was disbanded, but the military-civilian institutional linkages it forged lived on. These coalesced in the so-called military-industrial complex, identified as early as 1961 by President Dwight Eisenhower (formerly Supreme Commander of Allied Forces in Europe) in his Farewell Address.

Bush shrank into the background after the war, in part because he was hardcore analog, pathologically opposed to digital computing to the very end of his life. That makes it all the more ironic that the real blossoming of the ideas contained within “As We May Think” came about with the development of Hypertext and the World Wide Web.

Ted Nelson, credited with inventing the concept of Hypertext in the 1960s with his Project Xanadu, took Bush’s concept of the memex as his inspiration. One finds Bush’s fingerprints are all over Nelson’s 1965 essay Complex Information Processing: a file structure for the complex, the changing, and the indeterminate.

Ditto for American engineer Douglas Engelbart’s oN-Line System (NLS), shown at the 1968 Fall Joint Computer Conference in San Francisco and dubbed as “the mother of all demos.” Engelbart’s personal computer prototype demonstrated dynamic links between documents, information retrieval via mouse, word processing, and collaborative work on the same file.

English computer scientist Tim Berners-Lee, who invented the World Wide Web in 1989, explicitly acknowledged the influence of Bush’s memex on his work at a 1995 symposium at MIT, commemorating the 50th anniversary of “As We May Think.”

The memex concept of associative processes linking ideas, explored so fruitfully in the creation of the Internet, continues to find purchase. New “associative trails” for information storage, retrieval, and sharing are being blazed across the spectrum of digital tools, including CRM.

Not so long ago, corrupted, broken, or stolen hard drives were often a catastrophic loss—if we hadn’t backed up our files to an external drive, we would have lost all our work. And while that is sometimes still the case, these days cloud storage and web-based apps have taken a lot of the weight off our psyches. Similarly, with modern SaaS tools, we can glide between MacBook Air, crusty office Dell, iPhone, and iPad, working in the same suite with auto-save of anything we do presumed.

If Bush were still around, he would most definitely approve. Maybe then he would write a meta-update of his famous essay, centered around the future record keeping of record keeping.