Advanced Research Projects Agency

1. Receipt

In October 1962 a psychoacoustician walked into the Pentagon and was handed a chequebook. J.C.R. Licklider — “Lick” to everyone who mattered — had spent the fifties studying the ear at MIT and the decade’s end at BBN demonstrating what a person and a computer might do to one another’s thinking. ARPA, an agency conjured out of the Sputnik panic, gave him a new office with a deliberately dull name, Information Processing Techniques, and a latitude no funding body has enjoyed since. The office had no laboratories and hardly any staff. What it had was a mission Licklider had already written down: his 1960 paper “Man-Computer Symbiosis” proposed that the computer was not a calculator to be fed batch jobs overnight but a colleague to think with in real time. The office existed to make the sentence come true.

2. Method of operation

The method fitted on an index card and mattered as much as the money. Fund people, not projects. Choose them by taste, not by peer review. Pay lavishly, oversee minimally, and never make a first-rate mind write the sort of proposal a committee could love. The contracts went to laboratories that were already scenes — Project MAC at MIT, Engelbart’s augmentation crusade at SRI, Carnegie Mellon, Berkeley, Utah — and then the office did the one thing no patron had thought to do: it wired them together. Principal investigators were convened annually and made to know one another; graduate students circulated between the labs like a currency; tools and code travelled with them. Sputnik had made blank-cheque patronage of the improbable politically possible, and word of money that came with freedom moved through the field like gossip — a licence talent followed. The funded sites stopped being rivals for a budget and became one invisible college.

3. The intergalactic memorandum

The college had a founding joke. In April 1963 Licklider addressed a memorandum to the “Members and Affiliates of the Intergalactic Computer Network” — meaning the scattered labs on his books — and worried at a question that did not yet have hardware: when every site runs its own incompatible machine, how will any of them share what it builds? The joke was load-bearing. Symbiosis gave the labs one story to work inside; the memo sketched the story’s missing infrastructure. Meanwhile the office’s other instrument did its quiet work: the demo. A thing shown running — Sutherland’s Sketchpad, Engelbart’s screens — was at once proof, recruitment poster and budget justification, and the annual gatherings ran on them.

Fig. 1 — the funding topology

4. The wire

Succession kept the taste intact. Licklider handed the office to Ivan Sutherland, still in his twenties and already the author of Sketchpad; Sutherland to Robert Taylor, who sat in the Pentagon with three terminals wired to three incompatible time-sharing systems and concluded that the absurdity was fundable; Taylor to Larry Roberts, who engineered the answer. From 1966 the office bankrolled the ARPANET, and in the autumn of 1969 the first nodes came alive between the community’s own sites. What the network carried, at first, was the community itself — files, logins, argument, electronic mail. The product became the connective tissue: the labs had built a tool for sharing resources and found they had built the place where they now lived. Under Bob Kahn and his collaborators its protocols opened outward, until the network of networks needed a name of its own.

5. Disposition of file

The golden phase ended by mandate. The Mansfield Amendment of 1970 required defence research to show military relevance, and the later DARPA drifted from symbiosis toward applications and weapons; the blue sky narrowed year by year. Success finished what Congress began: the generation the office had trained dispersed — into Xerox PARC, where the personal computer was assembled from ARPA-bred parts, and into the departments and companies that staffed computing for decades. By then the legacy had already escaped the file: the internet, an ARPA-trained generation, and a model of visionary programme management every agency since has tried, and mostly failed, to copy. An office of a director and a few programme managers stands as arguably the highest-yield research patronage of the twentieth century — and the garden’s cleanest case of meta-scenius: a patron whose genius was to fund other people’s scenes, then wire the scenes together until they became one network.

6. Engines on record

7. Directors of record

8. Instruments of patronage

9. Assessment

10. References

1. [1]M. Mitchell Waldrop, The Dream Machine: J.C.R. Licklider and the Revolution That Made Computing Personal (2001). — Licklider’s vision and the IPTO story entire.
2. [2]Katie Hafner & Matthew Lyon, Where Wizards Stay Up Late: The Origins of the Internet (1996). — the building of the ARPANET.
3. [3]Janet Abbate, Inventing the Internet (1999). — the ARPANET-to-internet transition and its politics.
4. [4]Arthur L. Norberg & Judy E. O’Neill, Transforming Computer Technology: Information Processing for the Pentagon, 1962–1986 (1996). — the institutional history of the office and its funding practice.
5. [5]J.C.R. Licklider, “Man-Computer Symbiosis” (1960). — the founding text.

Keyed to the standard literature; edition-level verification pending.