As We May Think

The situation

In 1945 Vannevar Bush was probably the most powerful scientist in the United States. As director of the Office of Scientific Research and Development, he had coordinated the American scientific war effort, including radar, the proximity fuse and the early stages of the atomic bomb. His essay appeared in The Atlantic Monthly in July 1945, after the war in Europe had ended and weeks before the bombs fell on Hiroshima and Nagasaki.

The essay asks what scientists should do now. Bush’s answer is that they should turn to a problem science itself had created: the accumulation of knowledge faster than anyone could find or use it. Research was being published in enormous volume, and much of it was effectively lost because nobody could locate it when it mattered.

What it argues

Bush states the problem in a sentence that has been quoted ever since: “The summation of human experience is being expanded at a prodigious rate, and the means we use for threading through the consequent maze to the momentarily important item is the same as was used in the days of square-rigged ships.”

He then diagnoses why. Libraries index material by category, alphabetically or by subject, so that an item lives in one place and is found by descending through a hierarchy. “The human mind does not work that way. It operates by association.” When we think of one thing, we move to another connected to it, and then to another, along a web of trails built up by experience.

Bush’s proposal follows from this. He describes a machine he calls the memex, a desk with translucent screens, a keyboard and levers, storing books, papers and notes on microfilm. Its essential feature is that the user can join any two items together, building a trail. Later the user, or anyone the trail is shared with, can follow it, adding side trails and comments. Bush imagines that “wholly new forms of encyclopedias will appear, ready made with a mesh of associative trails running through them,” and that a new profession of trail blazers would find delight in establishing useful trails through the record.

Much of the essay is spent on hardware: dry photography, microfilm, voice-operated typewriters, cameras worn on the forehead. Some of it reads as quaint now. But the central idea is not the hardware. It is that knowledge should be organised the way people think, by association, and that individuals should be able to build and share the paths they make.

The idea underneath

Bush’s essay contains a theory of mind and a theory of knowledge. The theory of mind is associationist, in a tradition that runs back to John Locke and David Hume, who held that ideas are linked in the mind by resemblance, contiguity and cause. Bush gives it an engineering form: if the mind works by association, then tools for thought should too.

The theory of knowledge is social. Knowledge, in Bush’s picture, is not only what is stored in documents but the trails people make between them: their interpretations, connections and judgements. Those trails are usually lost when a person dies or forgets. The memex would make them durable and shareable. This anticipates what the essay on dispersed knowledge describes: much of what people know lies in how they connect things, not in the things themselves.

There is also an unspoken tension. Bush assumes that more access is better. The pharmakon tradition would ask what a memex takes away: perhaps the memory and synthesis that come from having to find and remember things oneself. The note-taking essay pursues that question.

How it was received and what followed

The essay was widely read and reprinted. Its most important readers came later. Douglas Engelbart read it as a young radar technician in 1945 and later said it shaped his life’s work, which led to the 1962 framework and the mouse-driven interface. Ted Nelson, who coined “hypertext” in the 1960s, built on the memex in designing Xanadu. Tim Berners-Lee’s 1989 proposal for the web pursued associative linking for a single organisation and then the world.

What none of them built was exactly the memex. The web links documents to documents, not personal trails, and its links belong to authors rather than readers. Bush’s idea of shareable personal trails has resurfaced repeatedly in note-taking tools, but it remains incompletely realised.

Reading it today as a decision-maker

The most useful thing about the essay is its method. Bush begins with a human problem, information overload, describes the experience a solution should give, and only then speculates about machinery. The machinery was wrong in nearly every detail; the experience was right. A product vision built this way survives changes in technology. One built around a specific technology does not.

The essay also shows the value of naming the mismatch between how a system stores things and how people use them. Bush saw that indexes followed the logic of storage while minds followed the logic of association. Many product failures stem from the same mismatch: a design that follows the database rather than the user.

Sources and limitations

Quotations are from The Atlantic’s archived text of the essay and were checked against widely reproduced versions; the passages quoted here are short. Details of Bush’s career are drawn from Zachary’s biography. The influence on Engelbart and Nelson is described in their own accounts, collected in Nyce and Kahn. The interpretation of Bush’s associationism as continuous with Locke and Hume is the author’s.